diff --git a/docs/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378.md b/docs/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378.md
index 5d504180..de6f7713 100644
--- a/docs/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378.md
+++ b/docs/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:58 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:45 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2023-03-15</td>
           <td>PLOS Computational Biology</td>
-          <td>3</td>
+          <td>5</td>
           <td>55</td>
         </tr>
     
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2022-06-01</td>
           <td>Nonlinear Dynamics</td>
-          <td>29</td>
-          <td>91</td>
+          <td>30</td>
+          <td>92</td>
         </tr>
     
         <tr id="Systems biology is a new discipline built upon the premise that an understanding of how cells and organisms carry out their functions cannot be gained by looking at cellular components in isolation. Instead, consideration of the interplay between the parts of systems is indispensable for analyzing, modeling, and predicting systems' behavior. Studying biological processes under this premise, systems biology combines experimental techniques and computational methods in order to construct predictive models. Both in building and utilizing models of biological systems, inverse problems arise at several occasions, for example, (i) when experimental time series and steady state data are used to construct biochemical reaction networks, (ii) when model parameters are identified that capture underlying mechanisms or (iii) when desired qualitative behavior such as bistability or limit cycle oscillations is engineered by proper choices of parameter combinations. In this paper we review principles of the modeling process in systems biology and illustrate the ill-posedness and regularization of parameter identification problems in that context. Furthermore, we discuss the methodology of qualitative inverse problems and demonstrate how sparsity enforcing regularization allows the determination of key reaction mechanisms underlying the qualitative behavior.">
@@ -126,6 +126,18 @@ hide:
           <td>48</td>
         </tr>
     
+        <tr id="Biological systems exhibit complex dynamics that differential equations can often adeptly represent. Ordinary differential equation models are widespread; until recently their construction has required extensive prior knowledge of the system. Machine learning methods offer alternative means of model construction: differential equation models can be learnt from data via model discovery using sparse identification of nonlinear dynamics (SINDy). However, SINDy struggles with realistic levels of biological noise and does not incorporate prior knowledge of the system. We propose a data-driven framework for model discovery and model selection using hybrid dynamical systems: partial models containing missing terms. Neural networks are used to approximate the unknown dynamics of a system, enabling the denoising the data while simultaneously learning the latent dynamics. Simulations from the fitted neural network are then used to infer models using SINDy. We show, via model selection, that model discovery in SINDy with hybrid dynamical systems outperforms alternative approaches. We find it possible to infer models correctly up to high levels of biological noise of different types. We demonstrate the potential to learn models from sparse, noisy data in application to a canonical cell state transition using data derived from single-cell transcriptomics. Overall, this approach provides a practical framework for model discovery in biology in cases where data are noisy and sparse, of particular utility when the underlying biological mechanisms are partially but incompletely known.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/596f63cbf87cf14a0e7859c17058a82b15c760ef" target='_blank'>Data-driven model discovery and model selection for noisy biological systems</a></td>
+          <td>
+            Xiaojun Wu, MeiLu McDermott, Adam L. Maclean
+          </td>
+          <td>2024-10-04</td>
+          <td>bioRxiv</td>
+          <td>0</td>
+          <td>19</td>
+        </tr>
+    
         <tr id="None">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/1626b462b65d4084a24fb31c4e4ca3fc212a307a" target='_blank'>Set-base dynamical parameter estimation and model invalidation for biochemical reaction networks</a></td>
@@ -135,7 +147,19 @@ hide:
           <td>2010-05-25</td>
           <td>BMC Systems Biology</td>
           <td>67</td>
-          <td>47</td>
+          <td>48</td>
+        </tr>
+    
+        <tr id="The combination of machine learning (ML) and sparsity-promoting techniques is enabling direct extraction of governing equations from data, revolutionizing computational modeling in diverse fields of science and engineering. The discovered dynamical models could be used to address challenges in climate science, neuroscience, ecology, finance, epidemiology, and beyond. However, most existing sparse identification methods for discovering dynamical systems treat the whole system as one without considering the interactions between subsystems. As a result, such models are not able to capture small changes in the emergent system behavior. To address this issue, we developed a new method called Sparse Identification of Nonlinear Dynamical Systems from Graph-structured data (SINDyG), which incorporates the network structure into sparse regression to identify model parameters that explain the underlying network dynamics. SINDyG discovers the governing equations of network dynamics while offering improvements in accuracy and model simplicity.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/4da518ae886702aed27abdd8f94fea3042ab9377" target='_blank'>Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems</a></td>
+          <td>
+            Mohammad Amin Basiri, Sina Khanmohammadi
+          </td>
+          <td>2024-09-02</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>3</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/0acd117521ef5aafb09fed02ab415523b330b058.md b/docs/recommendations/0acd117521ef5aafb09fed02ab415523b330b058.md
index bcbaecc3..5c8725d7 100644
--- a/docs/recommendations/0acd117521ef5aafb09fed02ab415523b330b058.md
+++ b/docs/recommendations/0acd117521ef5aafb09fed02ab415523b330b058.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:06 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:51 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2023-11-01</td>
           <td>Chaos</td>
-          <td>3</td>
+          <td>4</td>
           <td>11</td>
         </tr>
     
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
     
         <tr id="None">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
     
@@ -98,8 +98,8 @@ hide:
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>250</td>
-          <td>12</td>
+          <td>289</td>
+          <td>13</td>
         </tr>
     
         <tr id="Discovering the partial differential equations underlying spatio-temporal datasets from very limited and highly noisy observations is of paramount interest in many scientific fields. However, it remains an open question to know when model discovery algorithms based on sparse regression can actually recover the underlying physical processes. In this work, we show the design matrices used to infer the equations by sparse regression can violate the irrepresentability condition (IRC) of the Lasso, even when derived from analytical PDE solutions (i.e. without additional noise). Sparse regression techniques which can recover the true underlying model under violated IRC conditions are therefore required, leading to the introduction of the randomised adaptive Lasso. We show once the latter is integrated within the deep learning model discovery framework DeepMod, a wide variety of nonlinear and chaotic canonical PDEs can be recovered: (1) up to $\mathcal{O}(2)$ higher noise-to-sample ratios than state-of-the-art algorithms, (2) with a single set of hyperparameters, which paves the road towards truly automated model discovery.">
@@ -111,7 +111,7 @@ hide:
           <td>2021-06-22</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
     
         <tr id="We investigate the problem of learning an evolution equation directly from some given data. This work develops a learning algorithm to identify the terms in the underlying partial differential equations and to approximate the coefficients of the terms only using data. The algorithm uses sparse optimization in order to perform feature selection and parameter estimation. The features are data driven in the sense that they are constructed using nonlinear algebraic equations on the spatial derivatives of the data. Several numerical experiments show the proposed method's robustness to data noise and size, its ability to capture the true features of the data, and its capability of performing additional analytics. Examples include shock equations, pattern formation, fluid flow and turbulence, and oscillatory convection.">
@@ -123,7 +123,7 @@ hide:
           <td>2017-01-16</td>
           <td></td>
           <td>0</td>
-          <td>15</td>
+          <td>16</td>
         </tr>
     
         <tr id="We propose a regression method based upon group sparsity that is capable of discovering parametrized governing dynamical equations of motion of a given system by time series measurements. The method balances model complexity and regression accuracy by selecting a parsimonious model via Pareto analysis. This gives a promising new technique for disambiguating governing equations from simple parametric dependencies in physical, biological and engineering systems.">
@@ -135,7 +135,7 @@ hide:
           <td>2017-12-01</td>
           <td>2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)</td>
           <td>12</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b.md b/docs/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b.md
index e8832c4b..8c86caeb 100644
--- a/docs/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b.md
+++ b/docs/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:15 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:10 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2023-10-24</td>
           <td>ArXiv</td>
-          <td>6</td>
+          <td>9</td>
           <td>50</td>
         </tr>
     
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>54</td>
+          <td>1</td>
+          <td>55</td>
         </tr>
     
         <tr id="Accurate forecasting of multivariate time series is an extensively studied subject in finance, transportation, and computer science. Fully mining the correlation and causation between the variables in a multivariate time series exhibits noticeable results in improving the performance of a time series model. Recently, some models have explored the dependencies between variables through end-to-end graph structure learning without the need for predefined graphs. However, current models do not incorporate the trade-off between efficiency and flexibility and lack the guidance of domain knowledge in the design of graph structure learning algorithms. This paper alleviates the above issues by proposing Balanced Graph Structure Learning for Forecasting (BGSLF), a novel deep learning model that joins graph structure learning and forecasting. Technically, BGSLF leverages the spatial information into convolutional operations and extracts temporal dynamics using the diffusion convolutional recurrent network. The proposed framework balance the trade-off between efficiency and flexibility by introducing Multi-Graph Generation Network (MGN) and Graph Selection Module. In addition, a method named Smooth Sparse Unit (SSU) is designed to sparse the learned graph structures, which conforms to the sparse spatial correlations in the real world. Extensive experiments on four real-world datasets demonstrate that our model achieves state-of-the-art performances with minor trainable parameters. Code will be made publicly available.">
@@ -87,7 +87,7 @@ hide:
           <td>2022-01-24</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>10</td>
+          <td>11</td>
         </tr>
     
         <tr id="We study a recent class of models which uses graph neural networks (GNNs) to improve forecasting in multivariate time series. The core assumption behind these models is that there is a latent graph between the time series (nodes) that governs the evolution of the multivariate time series. By parameterizing a graph in a differentiable way, the models aim to improve forecasting quality. We compare four recent models of this class on the forecasting task. Further, we perform ablations to study their behavior under changing conditions, e.g., when disabling the graph-learning modules and providing the ground-truth relations instead. Based on our findings, we propose novel ways of combining the existing architectures.">
@@ -99,7 +99,7 @@ hide:
           <td>2021-09-10</td>
           <td>ArXiv</td>
           <td>12</td>
-          <td>46</td>
+          <td>47</td>
         </tr>
     
         <tr id="Multivariate time-series forecasting is a critical task for many applications, and graph time-series network is widely studied due to its capability to capture the spatial-temporal correlation simultaneously. However, most existing works focus more on learning with the explicit prior graph structure, while ignoring potential information from the implicit graph structure, yielding incomplete structure modeling. Some recent works attempts to learn the intrinsic or implicit graph structure directly, while lacking a way to combine explicit prior structure with implicit structure together. In this paper, we propose Regularized Graph Structure Learning (RGSL) model to incorporate both explicit prior structure and implicit structure together, and learn the forecasting deep networks along with the graph structure. RGSL consists of two innovative modules. First, we derive an implicit dense similarity matrix through node embedding, and learn the sparse graph structure using the Regularized Graph Generation (RGG) based on the Gumbel Softmax trick. Second, we propose a Laplacian Matrix Mixed-up Module (LM3) to fuse the explicit graph and implicit graph together. We conduct experiments on three real-word datasets. Results show that the proposed RGSL model outperforms existing graph forecasting algorithms with a notable margin, while learning meaningful graph structure simultaneously. Our code and models are made publicly available at https://github.com/alipay/RGSL.git.">
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2022-07-01</td>
           <td>ArXiv, DBLP</td>
-          <td>40</td>
-          <td>34</td>
+          <td>44</td>
+          <td>35</td>
         </tr>
     
         <tr id="Time series forecasting is an extensively studied subject in statistics, economics, and computer science. Exploration of the correlation and causation among the variables in a multivariate time series shows promise in enhancing the performance of a time series model. When using deep neural networks as forecasting models, we hypothesize that exploiting the pairwise information among multiple (multivariate) time series also improves their forecast. If an explicit graph structure is known, graph neural networks (GNNs) have been demonstrated as powerful tools to exploit the structure. In this work, we propose learning the structure simultaneously with the GNN if the graph is unknown. We cast the problem as learning a probabilistic graph model through optimizing the mean performance over the graph distribution. The distribution is parameterized by a neural network so that discrete graphs can be sampled differentiably through reparameterization. Empirical evaluations show that our method is simpler, more efficient, and better performing than a recently proposed bilevel learning approach for graph structure learning, as well as a broad array of forecasting models, either deep or non-deep learning based, and graph or non-graph based.">
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2021-01-18</td>
           <td>ArXiv</td>
-          <td>179</td>
-          <td>37</td>
+          <td>199</td>
+          <td>38</td>
         </tr>
     
         <tr id="Multi-variate time series forecasting is an important problem with a wide range of applications. Recent works model the relations between time-series as graphs and have shown that propagating information over the relation graph can improve time series forecasting. However, in many cases, relational information is not available or is noisy and reliable. Moreover, most works ignore the underlying uncertainty of time-series both for structure learning and deriving the forecasts resulting in the structure not capturing the uncertainty resulting in forecast distributions with poor uncertainty estimates. We tackle this challenge and introduce STOIC, that leverages stochastic correlations between time-series to learn underlying structure between time-series and to provide well-calibrated and accurate forecasts. Over a wide-range of benchmark datasets STOIC provides around 16% more accurate and 14% better-calibrated forecasts. STOIC also shows better adaptation to noise in data during inference and captures important and useful relational information in various benchmarks.">
diff --git a/docs/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b.md b/docs/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b.md
index 5aa37cd2..e92846a4 100644
--- a/docs/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b.md
+++ b/docs/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:17 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:12 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -44,41 +44,41 @@ hide:
     
         <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4fb78450650894091b0a55b5504a8b0a6f3dec37" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
           <td>
             Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>28</td>
+          <td>69</td>
           <td>18</td>
         </tr>
     
-        <tr id="This research examines the use of Large Language Models (LLMs) in predicting time series, with a specific focus on the LLMTIME model. Despite the established effectiveness of LLMs in tasks such as text generation, language translation, and sentiment analysis, this study highlights the key challenges that large language models encounter in the context of time series prediction. We assess the performance of LLMTIME across multiple datasets and introduce classical almost periodic functions as time series to gauge its effectiveness. The empirical results indicate that while large language models can perform well in zero-shot forecasting for certain datasets, their predictive accuracy diminishes notably when confronted with diverse time series data and traditional signals. The primary finding of this study is that the predictive capacity of LLMTIME, similar to other LLMs, significantly deteriorates when dealing with time series data that contain both periodic and trend components, as well as when the signal comprises complex frequency components.">
+        <tr id="Foundation models of time series have not been fully developed due to the limited availability of time series corpora and the underexploration of scalable pre-training. Based on the similar sequential formulation of time series and natural language, increasing research demonstrates the feasibility of leveraging large language models (LLM) for time series. Nevertheless, the inherent autoregressive property and decoder-only architecture of LLMs have not been fully considered, resulting in insufficient utilization of LLM abilities. To fully revitalize the general-purpose token transition and multi-step generation capability of large language models, we propose AutoTimes to repurpose LLMs as autoregressive time series forecasters, which projects time series into the embedding space of language tokens and autoregressively generates future predictions with arbitrary lengths. Compatible with any decoder-only LLMs, the consequent forecaster exhibits the flexibility of the lookback length and scalability with larger LLMs. Further, we formulate time series as prompts, extending the context for prediction beyond the lookback window, termed in-context forecasting. By introducing LLM-embedded textual timestamps, AutoTimes can utilize chronological information to align multivariate time series. Empirically, AutoTimes achieves state-of-the-art with 0.1% trainable parameters and over $5\times$ training/inference speedup compared to advanced LLM-based forecasters. Code is available at this repository: https://github.com/thuml/AutoTimes.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/11e794835a1f7d46560666d90c8eab3c24d84c05" target='_blank'>An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/d488445bb2bf6719bc48a4d39bd906116274abda" target='_blank'>AutoTimes: Autoregressive Time Series Forecasters via Large Language Models</a></td>
           <td>
-            Rui Cao, Qiao Wang
+            Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long
           </td>
-          <td>2024-08-09</td>
+          <td>2024-02-04</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>3</td>
+          <td>5</td>
+          <td>67</td>
         </tr>
     
-        <tr id="Foundation models of time series have not been fully developed due to the limited availability of time series corpora and the underexploration of scalable pre-training. Based on the similar sequential formulation of time series and natural language, increasing research demonstrates the feasibility of leveraging large language models (LLM) for time series. Nevertheless, the inherent autoregressive property and decoder-only architecture of LLMs have not been fully considered, resulting in insufficient utilization of LLM abilities. To further exploit the general-purpose token transition and multi-step generation ability of large language models, we propose AutoTimes to repurpose LLMs as autoregressive time series forecasters, which independently projects time series segments into the embedding space and autoregressively generates future predictions with arbitrary lengths. Compatible with any decoder-only LLMs, the consequent forecaster exhibits the flexibility of the lookback length and scalability of the LLM size. Further, we formulate time series as prompts, extending the context for prediction beyond the lookback window, termed in-context forecasting. By adopting textual timestamps as position embeddings, AutoTimes integrates multimodality for multivariate scenarios. Empirically, AutoTimes achieves state-of-the-art with 0.1% trainable parameters and over 5 times training/inference speedup compared to advanced LLM-based forecasters.">
+        <tr id="This research examines the use of Large Language Models (LLMs) in predicting time series, with a specific focus on the LLMTIME model. Despite the established effectiveness of LLMs in tasks such as text generation, language translation, and sentiment analysis, this study highlights the key challenges that large language models encounter in the context of time series prediction. We assess the performance of LLMTIME across multiple datasets and introduce classical almost periodic functions as time series to gauge its effectiveness. The empirical results indicate that while large language models can perform well in zero-shot forecasting for certain datasets, their predictive accuracy diminishes notably when confronted with diverse time series data and traditional signals. The primary finding of this study is that the predictive capacity of LLMTIME, similar to other LLMs, significantly deteriorates when dealing with time series data that contain both periodic and trend components, as well as when the signal comprises complex frequency components.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/d488445bb2bf6719bc48a4d39bd906116274abda" target='_blank'>AutoTimes: Autoregressive Time Series Forecasters via Large Language Models</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/11e794835a1f7d46560666d90c8eab3c24d84c05" target='_blank'>An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting</a></td>
           <td>
-            Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long
+            Rui Cao, Qiao Wang
           </td>
-          <td>2024-02-04</td>
+          <td>2024-08-09</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>66</td>
+          <td>0</td>
+          <td>3</td>
         </tr>
     
-        <tr id="Large language models (LLMs) are being applied to time series tasks, particularly time series forecasting. However, are language models actually useful for time series? After a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade the forecasting results -- in most cases the results even improved. We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and reveal that patching and attention structures perform similarly to state-of-the-art LLM-based forecasters.">
+        <tr id="Large language models (LLMs) are being applied to time series forecasting. But are language models actually useful for time series? In a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade forecasting performance -- in most cases, the results even improve! We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and find that patching and attention structures perform similarly to LLM-based forecasters.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/df0d604b8e8e3b2947d9865d735f204c08635012" target='_blank'>Are Language Models Actually Useful for Time Series Forecasting?</a></td>
           <td>
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2024-06-22</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>1</td>
+          <td>13</td>
+          <td>3</td>
         </tr>
     
         <tr id="Deep learning has contributed remarkably to the advancement of time series analysis. Still, deep models can encounter performance bottlenecks in real-world data-scarce scenarios, which can be concealed due to the performance saturation with small models on current benchmarks. Meanwhile, large models have demonstrated great powers in these scenarios through large-scale pre-training. Continuous progress has been achieved with the emergence of large language models, exhibiting unprecedented abilities such as few-shot generalization, scalability, and task generality, which are however absent in small deep models. To change the status quo of training scenario-specific small models from scratch, this paper aims at the early development of large time series models (LTSM). During pre-training, we curate large-scale datasets with up to 1 billion time points, unify heterogeneous time series into single-series sequence (S3) format, and develop the GPT-style architecture toward LTSMs. To meet diverse application needs, we convert forecasting, imputation, and anomaly detection of time series into a unified generative task. The outcome of this study is a Time Series Transformer (Timer), which is generative pre-trained by next token prediction and adapted to various downstream tasks with promising capabilities as an LTSM. Code and datasets are available at: https://github.com/thuml/Large-Time-Series-Model.">
@@ -98,8 +98,8 @@ hide:
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>4</td>
-          <td>66</td>
+          <td>21</td>
+          <td>67</td>
         </tr>
     
         <tr id="In this paper, we introduce TimeGPT, the first foundation model for time series, capable of generating accurate predictions for diverse datasets not seen during training. We evaluate our pre-trained model against established statistical, machine learning, and deep learning methods, demonstrating that TimeGPT zero-shot inference excels in performance, efficiency, and simplicity. Our study provides compelling evidence that insights from other domains of artificial intelligence can be effectively applied to time series analysis. We conclude that large-scale time series models offer an exciting opportunity to democratize access to precise predictions and reduce uncertainty by leveraging the capabilities of contemporary advancements in deep learning.">
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2023-10-05</td>
           <td>ArXiv</td>
-          <td>43</td>
-          <td>1</td>
+          <td>60</td>
+          <td>5</td>
         </tr>
     
         <tr id="Predicting future values in multivariate time series is vital across various domains. This work explores the use of large language models (LLMs) for this task. However, LLMs typically handle one-dimensional data. We introduce MultiCast, a zero-shot LLM-based approach for multivariate time series forecasting. It allows LLMs to receive multivariate time series as input, through three novel token multiplexing solutions that effectively reduce dimensionality while preserving key repetitive patterns. Additionally, a quantization scheme helps LLMs to better learn these patterns, while significantly reducing token use for practical applications. We showcase the performance of our approach in terms of RMSE and execution time against state-of-the-art approaches on three real-world datasets.">
@@ -134,8 +134,8 @@ hide:
           </td>
           <td>2024-02-16</td>
           <td>ArXiv</td>
-          <td>3</td>
-          <td>6</td>
+          <td>5</td>
+          <td>8</td>
         </tr>
     
         <tr id="Time series forecasting holds significant importance in many real-world dynamic systems and has been extensively studied. Unlike natural language process (NLP) and computer vision (CV), where a single large model can tackle multiple tasks, models for time series forecasting are often specialized, necessitating distinct designs for different tasks and applications. While pre-trained foundation models have made impressive strides in NLP and CV, their development in time series domains has been constrained by data sparsity. Recent studies have revealed that large language models (LLMs) possess robust pattern recognition and reasoning abilities over complex sequences of tokens. However, the challenge remains in effectively aligning the modalities of time series data and natural language to leverage these capabilities. In this work, we present Time-LLM, a reprogramming framework to repurpose LLMs for general time series forecasting with the backbone language models kept intact. We begin by reprogramming the input time series with text prototypes before feeding it into the frozen LLM to align the two modalities. To augment the LLM's ability to reason with time series data, we propose Prompt-as-Prefix (PaP), which enriches the input context and directs the transformation of reprogrammed input patches. The transformed time series patches from the LLM are finally projected to obtain the forecasts. Our comprehensive evaluations demonstrate that Time-LLM is a powerful time series learner that outperforms state-of-the-art, specialized forecasting models. Moreover, Time-LLM excels in both few-shot and zero-shot learning scenarios.">
@@ -146,7 +146,7 @@ hide:
           </td>
           <td>2023-10-03</td>
           <td>ArXiv</td>
-          <td>134</td>
+          <td>195</td>
           <td>9</td>
         </tr>
     
diff --git a/docs/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277.md b/docs/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277.md
index 98a85704..34a6a6be 100644
--- a/docs/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277.md
+++ b/docs/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:30 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:18 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -51,7 +51,7 @@ hide:
           <td>2024-05-23</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>6</td>
+          <td>7</td>
         </tr>
     
         <tr id="Large language models (LLMs) have been applied in many fields and have developed rapidly in recent years. As a classic machine learning task, time series forecasting has recently been boosted by LLMs. Recent works treat large language models as \emph{zero-shot} time series reasoners without further fine-tuning, which achieves remarkable performance. However, there are some unexplored research problems when applying LLMs for time series forecasting under the zero-shot setting. For instance, the LLMs' preferences for the input time series are less understood. In this paper, by comparing LLMs with traditional time series forecasting models, we observe many interesting properties of LLMs in the context of time series forecasting. First, our study shows that LLMs perform well in predicting time series with clear patterns and trends, but face challenges with datasets lacking periodicity. This observation can be explained by the ability of LLMs to recognize the underlying period within datasets, which is supported by our experiments. In addition, the input strategy is investigated, and it is found that incorporating external knowledge and adopting natural language paraphrases substantially improve the predictive performance of LLMs for time series. Overall, our study contributes insight into LLMs' advantages and limitations in time series forecasting under different conditions.">
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2024-02-16</td>
           <td>ArXiv</td>
-          <td>3</td>
-          <td>6</td>
+          <td>5</td>
+          <td>8</td>
         </tr>
     
         <tr id="Time-series forecasting (TSF) finds broad applications in real-world scenarios. Prompting off-the-shelf Large Language Models (LLMs) demonstrates strong zero-shot TSF capabilities while preserving computational efficiency. However, existing prompting methods oversimplify TSF as language next-token predictions, overlooking its dynamic nature and lack of integration with state-of-the-art prompt strategies such as Chain-of-Thought. Thus, we propose LSTPrompt, a novel approach for prompting LLMs in zero-shot TSF tasks. LSTPrompt decomposes TSF into short-term and long-term forecasting sub-tasks, tailoring prompts to each. LSTPrompt guides LLMs to regularly reassess forecasting mechanisms to enhance adaptability. Extensive evaluations demonstrate consistently better performance of LSTPrompt than existing prompting methods, and competitive results compared to foundation TSF models.">
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2024-02-25</td>
           <td>ArXiv</td>
-          <td>6</td>
-          <td>7</td>
+          <td>11</td>
+          <td>8</td>
         </tr>
     
         <tr id="This paper presents a new perspective on time series forecasting. In existing time series forecasting methods, the models take a sequence of numerical values as input and yield numerical values as output. The existing SOTA models are largely based on the Transformer architecture, modified with multiple encoding mechanisms to incorporate the context and semantics around the historical data. Inspired by the successes of pre-trained language foundation models, we pose a question about whether these models can also be adapted to solve time-series forecasting. Thus, we propose a new forecasting paradigm: prompt-based time series forecasting (PromptCast). In this novel task, the numerical input and output are transformed into prompts and the forecasting task is framed in a sentence-to-sentence manner, making it possible to directly apply language models for forecasting purposes. To support and facilitate the research of this task, we also present a large-scale dataset (PISA) that includes three real-world forecasting scenarios. We evaluate different SOTA numerical-based forecasting methods and language generation models. The benchmark results with various forecasting settings demonstrate the proposed PromptCast with language generation models is a promising research direction. Additionally, in comparison to conventional numerical-based forecasting, PromptCast shows a much better generalization ability under the zero-shot setting.">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2022-09-20</td>
           <td>IEEE Transactions on Knowledge and Data Engineering</td>
-          <td>63</td>
+          <td>80</td>
           <td>17</td>
         </tr>
     
@@ -99,7 +99,7 @@ hide:
           <td>2024-06-20</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>23</td>
+          <td>24</td>
         </tr>
     
         <tr id="Time series data are ubiquitous across various domains, making time series analysis critically important. Traditional time series models are task-specific, featuring singular functionality and limited generalization capacity. Recently, large language foundation models have unveiled their remarkable capabilities for cross-task transferability, zero-shot/few-shot learning, and decision-making explainability. This success has sparked interest in the exploration of foundation models to solve multiple time series challenges simultaneously. There are two main research lines, namely pre-training foundation models from scratch for time series and adapting large language foundation models for time series. They both contribute to the development of a unified model that is highly generalizable, versatile, and comprehensible for time series analysis. This survey offers a 3E analytical framework for comprehensive examination of related research. Specifically, we examine existing works from three dimensions, namely Effectiveness, Efficiency and Explainability. In each dimension, we focus on discussing how related works devise tailored solution by considering unique challenges in the realm of time series. Furthermore, we provide a domain taxonomy to help followers keep up with the domain-specific advancements. In addition, we introduce extensive resources to facilitate the field's development, including datasets, open-source, time series libraries. A GitHub repository is also maintained for resource updates (https://github.com/start2020/Awesome-TimeSeries-LLM-FM).">
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2024-05-03</td>
           <td>ArXiv</td>
-          <td>4</td>
+          <td>7</td>
           <td>47</td>
         </tr>
     
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2024-05-23</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>1</td>
+          <td>2</td>
+          <td>2</td>
         </tr>
     
         <tr id="Multivariate time-series forecasting is vital in various domains, e.g., economic planning and weather prediction. Deep train-from-scratch models have exhibited effective performance yet require large amounts of data, which limits real-world applicability. Recently, researchers have leveraged the representation learning transferability of pre-trained Large Language Models (LLMs) to handle limited non-linguistic datasets effectively. However, incorporating LLMs with time-series data presents challenges of limited adaptation due to different compositions between time-series and linguistic data, and the inability to process multi-scale temporal information. To tackle these challenges, we propose LLM4TS, a framework for time-series forecasting with pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the \textit{time-series alignment} stage to align LLMs with the nuances of time-series data, and the \textit{forecasting fine-tuning} stage for downstream time-series forecasting tasks. Furthermore, our framework features a novel two-level aggregation method that integrates multi-scale temporal data within pre-trained LLMs, enhancing their ability to interpret time-specific information. In experiments across 7 time-series forecasting datasets, LLM4TS is superior to existing state-of-the-art methods compared with trained-from-scratch models in full-shot scenarios, and also achieves an average improvement of 6.84% in MSE in few-shot scenarios. In addition, evaluations compared with different self-supervised learning approaches highlight LLM4TS's effectiveness with representation learning in forecasting tasks.">
@@ -134,32 +134,20 @@ hide:
           </td>
           <td>2023-08-16</td>
           <td>ArXiv</td>
-          <td>16</td>
-          <td>2</td>
+          <td>22</td>
+          <td>3</td>
         </tr>
     
-        <tr id="Large language models (LLMs) are being applied to time series tasks, particularly time series forecasting. However, are language models actually useful for time series? After a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade the forecasting results -- in most cases the results even improved. We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and reveal that patching and attention structures perform similarly to state-of-the-art LLM-based forecasters.">
+        <tr id="Multi-modal large language models (MLLMs) have enabled numerous advances in understanding and reasoning in domains like vision, but we have not yet seen this broad success for time-series. Although prior works on time-series MLLMs have shown promising performance in time-series forecasting, very few works show how an LLM could be used for time-series reasoning in natural language. We propose a novel multi-modal time-series LLM approach that learns generalizable information across various domains with powerful zero-shot performance. First, we train a lightweight time-series encoder on top of an LLM to directly extract time-series information. Then, we fine-tune our model with chain-of-thought augmented time-series tasks to encourage the model to generate reasoning paths. We show that our model learns a latent representation that reflects specific time-series features (e.g. slope, frequency), as well as outperforming GPT-4o on a set of zero-shot reasoning tasks on a variety of domains.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/df0d604b8e8e3b2947d9865d735f204c08635012" target='_blank'>Are Language Models Actually Useful for Time Series Forecasting?</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/e2b1ba7e73ca4ce0abef8c87f2c266e502b3b7b4" target='_blank'>Towards Time Series Reasoning with LLMs</a></td>
           <td>
-            Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen
+            Winnie Chow, Lauren Gardiner, Haraldur T. Hallgr'imsson, Maxwell A. Xu, Shirley You Ren
           </td>
-          <td>2024-06-22</td>
+          <td>2024-09-17</td>
           <td>ArXiv</td>
+          <td>2</td>
           <td>1</td>
-          <td>1</td>
-        </tr>
-    
-        <tr id="Motivated by recent advances in large language models for Natural Language Processing (NLP), we design a time-series foundation model for forecasting whose out-of-the-box zero-shot performance on a variety of public datasets comes close to the accuracy of state-of-the-art supervised forecasting models for each individual dataset. Our model is based on pretraining a patched-decoder style attention model on a large time-series corpus, and can work well across different forecasting history lengths, prediction lengths and temporal granularities.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/f45f85fa1beaa795c24c4ff86f1f2deece72252f" target='_blank'>A decoder-only foundation model for time-series forecasting</a></td>
-          <td>
-            Abhimanyu Das, Weihao Kong, Rajat Sen, Yichen Zhou
-          </td>
-          <td>2023-10-14</td>
-          <td>ArXiv</td>
-          <td>51</td>
-          <td>14</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f.md b/docs/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f.md
index ad5f46c3..13d0161f 100644
--- a/docs/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f.md
+++ b/docs/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:17 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:13 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2024-05-22</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>2</td>
+          <td>1</td>
+          <td>6</td>
         </tr>
     
         <tr id="Graph Neural Networks (GNNs) have shown promising potential in graph representation learning. The majority of GNNs define a local message-passing mechanism, propagating information over the graph by stacking multiple layers. These methods, however, are known to suffer from two major limitations: over-squashing and poor capturing of long-range dependencies. Recently, Graph Transformers (GTs) emerged as a powerful alternative to Message-Passing Neural Networks (MPNNs). GTs, however, have quadratic computational cost, lack inductive biases on graph structures, and rely on complex Positional/Structural Encodings (SE/PE). In this paper, we show that while Transformers, complex message-passing, and SE/PE are sufficient for good performance in practice, neither is necessary. Motivated by the recent success of State Space Models (SSMs), such as Mamba, we present Graph Mamba Networks (GMNs), a general framework for a new class of GNNs based on selective SSMs. We discuss and categorize the new challenges when adapting SSMs to graph-structured data, and present four required and one optional steps to design GMNs, where we choose (1) Neighborhood Tokenization, (2) Token Ordering, (3) Architecture of Bidirectional Selective SSM Encoder, (4) Local Encoding, and dispensable (5) PE and SE. We further provide theoretical justification for the power of GMNs. Experiments demonstrate that despite much less computational cost, GMNs attain an outstanding performance in long-range, small-scale, large-scale, and heterophilic benchmark datasets.">
@@ -62,11 +62,11 @@ hide:
           </td>
           <td>2024-02-13</td>
           <td>ArXiv, DBLP</td>
-          <td>33</td>
+          <td>39</td>
           <td>9</td>
         </tr>
     
-        <tr id="Message-passing graph neural networks (GNNs), while excelling at capturing local relationships, often struggle with long-range dependencies on graphs. Conversely, graph transformers (GTs) enable information exchange between all nodes but oversimplify the graph structure by treating them as a set of fixed-length vectors. This work proposes a novel architecture, NeuralWalker, that overcomes the limitations of both methods by combining random walks with message passing. NeuralWalker achieves this by treating random walks as sequences, allowing for the application of recent advances in sequence models in order to capture long-range dependencies within these walks. Based on this concept, we propose a framework that offers (1) more expressive graph representations through random walk sequences, (2) the ability to utilize any sequence model for capturing long-range dependencies, and (3) the flexibility by integrating various GNN and GT architectures. Our experimental evaluations demonstrate that NeuralWalker achieves significant performance improvements on 19 graph and node benchmark datasets, notably outperforming existing methods by up to 13% on the PascalVoc-SP and COCO-SP datasets. Code is available at https://github.com/BorgwardtLab/NeuralWalker.">
+        <tr id="Message-passing graph neural networks (GNNs) excel at capturing local relationships but struggle with long-range dependencies in graphs. In contrast, graph transformers (GTs) enable global information exchange but often oversimplify the graph structure by representing graphs as sets of fixed-length vectors. This work introduces a novel architecture that overcomes the shortcomings of both approaches by combining the long-range information of random walks with local message passing. By treating random walks as sequences, our architecture leverages recent advances in sequence models to effectively capture long-range dependencies within these walks. Based on this concept, we propose a framework that offers (1) more expressive graph representations through random walk sequences, (2) the ability to utilize any sequence model for capturing long-range dependencies, and (3) the flexibility by integrating various GNN and GT architectures. Our experimental evaluations demonstrate that our approach achieves significant performance improvements on 19 graph and node benchmark datasets, notably outperforming existing methods by up to 13\% on the PascalVoc-SP and COCO-SP datasets. The code is available at https://github.com/BorgwardtLab/NeuralWalker.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/0702ac9db3ea20810ffdb9f35bde85cdfaf27a3d" target='_blank'>Learning Long Range Dependencies on Graphs via Random Walks</a></td>
           <td>
@@ -74,11 +74,11 @@ hide:
           </td>
           <td>2024-06-05</td>
           <td>ArXiv</td>
-          <td>0</td>
+          <td>2</td>
           <td>8</td>
         </tr>
     
-        <tr id="Machine learning on graphs has recently found extensive applications across domains. However, the commonly used Message Passing Neural Networks (MPNNs) suffer from limited expressive power and struggle to capture long-range dependencies. Graph transformers offer a strong alternative due to their global attention mechanism, but they come with great computational overheads, especially for large graphs. In recent years, State Space Models (SSMs) have emerged as a compelling approach to replace full attention in transformers to model sequential data. It blends the strengths of RNNs and CNNs, offering a) efficient computation, b) the ability to capture long-range dependencies, and c) good generalization across sequences of various lengths. However, extending SSMs to graph-structured data presents unique challenges due to the lack of canonical node ordering in graphs. In this work, we propose Graph State Space Convolution (GSSC) as a principled extension of SSMs to graph-structured data. By leveraging global permutation-equivariant set aggregation and factorizable graph kernels that rely on relative node distances as the convolution kernels, GSSC preserves all three advantages of SSMs. We demonstrate the provably stronger expressiveness of GSSC than MPNNs in counting graph substructures and show its effectiveness across 10 real-world, widely used benchmark datasets, where GSSC achieves best results on 7 out of 10 datasets with all significant improvements compared to the state-of-the-art baselines and second-best results on the other 3 datasets. Our findings highlight the potential of GSSC as a powerful and scalable model for graph machine learning. Our code is available at https://github.com/Graph-COM/GSSC.">
+        <tr id="Machine learning on graphs has recently found extensive applications across domains. However, the commonly used Message Passing Neural Networks (MPNNs) suffer from limited expressive power and struggle to capture long-range dependencies. Graph transformers offer a strong alternative due to their global attention mechanism, but they come with great computational overheads, especially for large graphs. In recent years, State Space Models (SSMs) have emerged as a compelling approach to replace full attention in transformers to model sequential data. It blends the strengths of RNNs and CNNs, offering a) efficient computation, b) the ability to capture long-range dependencies, and c) good generalization across sequences of various lengths. However, extending SSMs to graph-structured data presents unique challenges due to the lack of canonical node ordering in graphs. In this work, we propose Graph State Space Convolution (GSSC) as a principled extension of SSMs to graph-structured data. By leveraging global permutation-equivariant set aggregation and factorizable graph kernels that rely on relative node distances as the convolution kernels, GSSC preserves all three advantages of SSMs. We demonstrate the provably stronger expressiveness of GSSC than MPNNs in counting graph substructures and show its effectiveness across 11 real-world, widely used benchmark datasets. GSSC achieves the best results on 6 out of 11 datasets with all significant improvements compared to the state-of-the-art baselines and second-best results on the other 5 datasets. Our findings highlight the potential of GSSC as a powerful and scalable model for graph machine learning. Our code is available at https://github.com/Graph-COM/GSSC.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/93a9b7940e902f989e27f94061d3b699a2fc4d13" target='_blank'>What Can We Learn from State Space Models for Machine Learning on Graphs?</a></td>
           <td>
@@ -90,6 +90,18 @@ hide:
           <td>3</td>
         </tr>
     
+        <tr id="Graph representation learning (GRL) is fundamental in multi-graph applications like molecular property prediction. Graph neural networks (GNNs) have emerged as a popular method for GRL. However, existing GRL methods primarily focus on designing GNN models with enhanced expressiveness while overlooking the memory efficiency of algorithms during training. The memory inefficiency problem is caused by a contextual constraint imposed on node representations, which requires each node to be context-dependent on its input graph. In this paper, we propose a novel method, called context sketching (COS), for memory-efficient graph representation learning in multi-graph scenarios. We first formally define the contextual constraint based on the enclosed $\infty-$hop subgraphs of nodes. Subsequently, we propose to relax the original contextual constraint by requiring each node to be context-dependent on its enclosed $k-$hop subgraph $(k \ll \infty)$ which is a contextual sketch of the enclosed $\infty$-hop subgraph. Lastly, we prove that COS constructs an optimal solution to a memory-related objective associated with graph coarsening. Experiments on four widely used benchmark datasets demonstrate that COS can reduce the memory footprint of baselines by a large margin with almost no accuracy loss.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/54288b7e9ce29a338f2b11981c6b84f4f0cb9a50" target='_blank'>Context Sketching for Memory-efficient Graph Representation Learning</a></td>
+          <td>
+            Kai-Lang Yao, Wusuo Li
+          </td>
+          <td>2023-12-01</td>
+          <td>2023 IEEE International Conference on Data Mining (ICDM)</td>
+          <td>0</td>
+          <td>4</td>
+        </tr>
+    
         <tr id="Graph Neural Networks (GNNs) have shown impressive performance in graph representation learning, but they face challenges in capturing long-range dependencies due to their limited expressive power. To address this, Graph Transformers (GTs) were introduced, utilizing self-attention mechanism to effectively model pairwise node relationships. Despite their advantages, GTs suffer from quadratic complexity w.r.t. the number of nodes in the graph, hindering their applicability to large graphs. In this work, we present Graph-Enhanced Contextual Operator (GECO), a scalable and effective alternative to GTs that leverages neighborhood propagation and global convolutions to effectively capture local and global dependencies in quasilinear time. Our study on synthetic datasets reveals that GECO reaches 169x speedup on a graph with 2M nodes w.r.t. optimized attention. Further evaluations on diverse range of benchmarks showcase that GECO scales to large graphs where traditional GTs often face memory and time limitations. Notably, GECO consistently achieves comparable or superior quality compared to baselines, improving the SOTA up to 4.5%, and offering a scalable and effective solution for large-scale graph learning.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/48052c9ebe066b9fcd653897dabf18582ec7e7fb" target='_blank'>A Scalable and Effective Alternative to Graph Transformers</a></td>
@@ -110,7 +122,7 @@ hide:
           </td>
           <td>2022-10-08</td>
           <td>ArXiv</td>
-          <td>59</td>
+          <td>63</td>
           <td>19</td>
         </tr>
     
@@ -126,18 +138,6 @@ hide:
           <td>28</td>
         </tr>
     
-        <tr id="Recent works have demonstrated the benefits of capturing long-distance dependency in graphs by deeper graph neural networks (GNNs). But deeper GNNs suffer from the long-lasting scalability challenge due to the neighborhood explosion problem in large-scale graphs. In this work, we propose to capture long-distance dependency in graphs by shallower models instead of deeper models, which leads to a much more efficient model, LazyGNN, for graph representation learning. Moreover, we demonstrate that LazyGNN is compatible with existing scalable approaches (such as sampling methods) for further accelerations through the development of mini-batch LazyGNN. Comprehensive experiments demonstrate its superior prediction performance and scalability on large-scale benchmarks. The implementation of LazyGNN is available at https://github.com/RXPHD/Lazy_GNN.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/5ca04cde78542b8d58c4c5b2db91ddd65b56d917" target='_blank'>LazyGNN: Large-Scale Graph Neural Networks via Lazy Propagation</a></td>
-          <td>
-            Rui Xue, Haoyu Han, MohamadAli Torkamani, Jian Pei, Xiaorui Liu
-          </td>
-          <td>2023-02-03</td>
-          <td>ArXiv</td>
-          <td>9</td>
-          <td>7</td>
-        </tr>
-    
   </tbody>
   <tfoot>
     <tr>
diff --git a/docs/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68.md b/docs/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68.md
index fa1886ae..d2250fd8 100644
--- a/docs/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68.md
+++ b/docs/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:34 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:28 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -49,8 +49,8 @@ hide:
             Andreas Doerr, Christian Daniel, Martin Schiegg, D. Nguyen-Tuong, S. Schaal, Marc Toussaint, Sebastian Trimpe
           </td>
           <td>2018-01-31</td>
-          <td>ArXiv, DBLP, MAG</td>
-          <td>111</td>
+          <td>ArXiv, MAG, DBLP</td>
+          <td>113</td>
           <td>93</td>
         </tr>
     
@@ -66,6 +66,18 @@ hide:
           <td>40</td>
         </tr>
     
+        <tr id="World modelling is essential for understanding and predicting the dynamics of complex systems by learning both spatial and temporal dependencies. However, current frameworks, such as Transformers and selective state-space models like Mambas, exhibit limitations in efficiently encoding spatial and temporal structures, particularly in scenarios requiring long-term high-dimensional sequence modelling. To address these issues, we propose a novel recurrent framework, the \textbf{FACT}ored \textbf{S}tate-space (\textbf{FACTS}) model, for spatial-temporal world modelling. The FACTS framework constructs a graph-structured memory with a routing mechanism that learns permutable memory representations, ensuring invariance to input permutations while adapting through selective state-space propagation. Furthermore, FACTS supports parallel computation of high-dimensional sequences. We empirically evaluate FACTS across diverse tasks, including multivariate time series forecasting and object-centric world modelling, demonstrating that it consistently outperforms or matches specialised state-of-the-art models, despite its general-purpose world modelling design.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/ec840755867d3f5cf175cb57de963f042297f4ef" target='_blank'>FACTS: A Factored State-Space Framework For World Modelling</a></td>
+          <td>
+            Li Nanbo, Firas Laakom, Yucheng Xu, Wenyi Wang, Jurgen Schmidhuber
+          </td>
+          <td>2024-10-28</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>7</td>
+        </tr>
+    
         <tr id="Time series with long-term structure arise in a variety of contexts and capturing this temporal structure is a critical challenge in time series analysis for both inference and forecasting settings. Traditionally, state space models have been successful in providing uncertainty estimates of trajectories in the latent space. More recently, deep learning, attention-based approaches have achieved state of the art performance for sequence modeling, though often require large amounts of data and parameters to do so. We propose Stanza, a nonlinear, non-stationary state space model as an intermediate approach to fill the gap between traditional models and modern deep learning approaches for complex time series. Stanza strikes a balance between competitive forecasting accuracy and probabilistic, interpretable inference for highly structured time series. In particular, Stanza achieves forecasting accuracy competitive with deep LSTMs on real-world datasets, especially for multi-step ahead forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/9c619e58d073772ff298228d47afcab625c7f37a" target='_blank'>Stanza: A Nonlinear State Space Model for Probabilistic Inference in Non-Stationary Time Series</a></td>
@@ -75,7 +87,7 @@ hide:
           <td>2020-06-11</td>
           <td>ArXiv</td>
           <td>6</td>
-          <td>31</td>
+          <td>32</td>
         </tr>
     
         <tr id="Outstanding achievements of graph neural networks for spatiotemporal time series analysis show that relational constraints introduce an effective inductive bias into neural forecasting architectures. Often, however, the relational information characterizing the underlying data-generating process is unavailable and the practitioner is left with the problem of inferring from data which relational graph to use in the subsequent processing stages. We propose novel, principled - yet practical - probabilistic score-based methods that learn the relational dependencies as distributions over graphs while maximizing end-to-end the performance at task. The proposed graph learning framework is based on consolidated variance reduction techniques for Monte Carlo score-based gradient estimation, is theoretically grounded, and, as we show, effective in practice. In this paper, we focus on the time series forecasting problem and show that, by tailoring the gradient estimators to the graph learning problem, we are able to achieve state-of-the-art performance while controlling the sparsity of the learned graph and the computational scalability. We empirically assess the effectiveness of the proposed method on synthetic and real-world benchmarks, showing that the proposed solution can be used as a stand-alone graph identification procedure as well as a graph learning component of an end-to-end forecasting architecture.">
@@ -86,7 +98,7 @@ hide:
           </td>
           <td>2022-05-26</td>
           <td>J. Mach. Learn. Res.</td>
-          <td>11</td>
+          <td>14</td>
           <td>50</td>
         </tr>
     
@@ -94,7 +106,7 @@ hide:
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/05cd3afc8208f1c0dd61b09a90f35dd42497e175" target='_blank'>Multi-Knowledge Fusion Network for Time Series Representation Learning</a></td>
           <td>
-            Sagar Srinivas Sakhinana, Shivam Gupta, Krishna Sai Sudhir Aripirala, Venkataramana Runkana
+            Sakhinana Sagar Srinivas, Shivam Gupta, Krishna Sai Sudhir Aripirala, Venkataramana Runkana
           </td>
           <td>2024-08-22</td>
           <td>ArXiv</td>
@@ -102,44 +114,28 @@ hide:
           <td>3</td>
         </tr>
     
-        <tr id="Over the past few years, research on deep graph learning has shifted from static graphs to temporal graphs in response to real-world complex systems that exhibit dynamic behaviors. In practice, temporal graphs are formalized as an ordered sequence of static graph snapshots observed at discrete time points. Sequence models such as RNNs or Transformers have long been the predominant backbone networks for modeling such temporal graphs. Yet, despite the promising results, RNNs struggle with long-range dependencies, while transformers are burdened by quadratic computational complexity. Recently, state space models (SSMs), which are framed as discretized representations of an underlying continuous-time linear dynamical system, have garnered substantial attention and achieved breakthrough advancements in independent sequence modeling. In this work, we undertake a principled investigation that extends SSM theory to temporal graphs by integrating structural information into the online approximation objective via the adoption of a Laplacian regularization term. The emergent continuous-time system introduces novel algorithmic challenges, thereby necessitating our development of GraphSSM, a graph state space model for modeling the dynamics of temporal graphs. Extensive experimental results demonstrate the effectiveness of our GraphSSM framework across various temporal graph benchmarks.">
+        <tr id="Real-world dynamical systems often consist of multiple stochastic subsystems that interact with each other. Modeling and forecasting the behavior of such dynamics are generally not easy, due to the inherent hardness in understanding the complicated interactions and evolutions of their constituents. This paper introduces the relational state-space model (R-SSM), a sequential hierarchical latent variable model that makes use of graph neural networks (GNNs) to simulate the joint state transitions of multiple correlated objects. By letting GNNs cooperate with SSM, R-SSM provides a flexible way to incorporate relational information into the modeling of multi-object dynamics. We further suggest augmenting the model with normalizing flows instantiated for vertex-indexed random variables and propose two auxiliary contrastive objectives to facilitate the learning. The utility of R-SSM is empirically evaluated on synthetic and real time series datasets.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/919e5db29c7b7be4468b975eb4c0fa4a543165fc" target='_blank'>State Space Models on Temporal Graphs: A First-Principles Study</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/7a1e5377b08489c2969f73c56efc557e34f578e1" target='_blank'>Relational State-Space Model for Stochastic Multi-Object Systems</a></td>
           <td>
-            Jintang Li, Ruofan Wu, Xinzhou Jin, Boqun Ma, Liang Chen, Zibin Zheng
+            Fan Yang, Ling Chen, Fan Zhou, Yusong Gao, Wei Cao
           </td>
-          <td>2024-06-03</td>
+          <td>2020-01-13</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>11</td>
+          <td>8</td>
+          <td>60</td>
         </tr>
     
-        <tr id="Time series modeling is a well-established problem, which often requires that methods (1) expressively represent complicated dependencies, (2) forecast long horizons, and (3) efficiently train over long sequences. State-space models (SSMs) are classical models for time series, and prior works combine SSMs with deep learning layers for efficient sequence modeling. However, we find fundamental limitations with these prior approaches, proving their SSM representations cannot express autoregressive time series processes. We thus introduce SpaceTime, a new state-space time series architecture that improves all three criteria. For expressivity, we propose a new SSM parameterization based on the companion matrix -- a canonical representation for discrete-time processes -- which enables SpaceTime's SSM layers to learn desirable autoregressive processes. For long horizon forecasting, we introduce a"closed-loop"variation of the companion SSM, which enables SpaceTime to predict many future time-steps by generating its own layer-wise inputs. For efficient training and inference, we introduce an algorithm that reduces the memory and compute of a forward pass with the companion matrix. With sequence length $\ell$ and state-space size $d$, we go from $\tilde{O}(d \ell)$ na\"ively to $\tilde{O}(d + \ell)$. In experiments, our contributions lead to state-of-the-art results on extensive and diverse benchmarks, with best or second-best AUROC on 6 / 7 ECG and speech time series classification, and best MSE on 14 / 16 Informer forecasting tasks. Furthermore, we find SpaceTime (1) fits AR($p$) processes that prior deep SSMs fail on, (2) forecasts notably more accurately on longer horizons than prior state-of-the-art, and (3) speeds up training on real-world ETTh1 data by 73% and 80% relative wall-clock time over Transformers and LSTMs.">
+        <tr id="Over the past few years, research on deep graph learning has shifted from static graphs to temporal graphs in response to real-world complex systems that exhibit dynamic behaviors. In practice, temporal graphs are formalized as an ordered sequence of static graph snapshots observed at discrete time points. Sequence models such as RNNs or Transformers have long been the predominant backbone networks for modeling such temporal graphs. Yet, despite the promising results, RNNs struggle with long-range dependencies, while transformers are burdened by quadratic computational complexity. Recently, state space models (SSMs), which are framed as discretized representations of an underlying continuous-time linear dynamical system, have garnered substantial attention and achieved breakthrough advancements in independent sequence modeling. In this work, we undertake a principled investigation that extends SSM theory to temporal graphs by integrating structural information into the online approximation objective via the adoption of a Laplacian regularization term. The emergent continuous-time system introduces novel algorithmic challenges, thereby necessitating our development of GraphSSM, a graph state space model for modeling the dynamics of temporal graphs. Extensive experimental results demonstrate the effectiveness of our GraphSSM framework across various temporal graph benchmarks.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/a7d68b1702af08ce4dbbf2cd0b083e744ae5c6be" target='_blank'>Effectively Modeling Time Series with Simple Discrete State Spaces</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/919e5db29c7b7be4468b975eb4c0fa4a543165fc" target='_blank'>State Space Models on Temporal Graphs: A First-Principles Study</a></td>
           <td>
-            Michael Zhang, Khaled Kamal Saab, Michael Poli, Tri Dao, Karan Goel, Christopher Ré
+            Jintang Li, Ruofan Wu, Xinzhou Jin, Boqun Ma, Liang Chen, Zibin Zheng
           </td>
-          <td>2023-03-16</td>
+          <td>2024-06-03</td>
           <td>ArXiv</td>
-          <td>30</td>
-          <td>44</td>
-        </tr>
-    
-        <tr id="
- 
- Gaussian state space models have been used for decades as generative models of sequential data. They admit an intuitive probabilistic interpretation, have a simple functional form, and enjoy widespread adoption. We introduce a unified algorithm to efficiently learn a broad class of linear and non-linear state space models, including variants where the emission and transition distributions are modeled by deep neural networks. Our learning algorithm simultaneously learns a compiled inference network and the generative model, leveraging a structured variational approximation parameterized by recurrent neural networks to mimic the posterior distribution. We apply the learning algorithm to both synthetic and real-world datasets, demonstrating its scalability and versatility. We find that using the structured approximation to the posterior results in models with significantly higher held-out likelihood.
- 
-">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2af17f153e3fd71e15db9216b972aef222f46617" target='_blank'>Structured Inference Networks for Nonlinear State Space Models</a></td>
-          <td>
-            R. G. Krishnan, Uri Shalit, D. Sontag
-          </td>
-          <td>2016-09-30</td>
-          <td>ArXiv, DBLP, MAG</td>
-          <td>429</td>
-          <td>48</td>
+          <td>1</td>
+          <td>12</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196.md b/docs/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196.md
index fa0f6c6f..cf399b69 100644
--- a/docs/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196.md
+++ b/docs/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:04 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:47 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2021-10-18</td>
           <td>ArXiv</td>
-          <td>18</td>
-          <td>50</td>
+          <td>21</td>
+          <td>51</td>
         </tr>
     
         <tr id="In this paper we propose a model reduction and identification approach for multilinear dynamical system (MLDS) driven by noise. Compared to standard linear dynamical system based approaches which fit vector or matrix models to tensor time series, MLDS provides more natural, compact and accurate representation of tensorial data with fewer model parameters. The proposed algorithm for identifying MLDS parameters employs techniques from multilinear subspace learning: mulilinear Principal Component Analysis and multilinear regression. In addition compact array normal distribution is used to represent and estimate model error and output noise.We illustrate the benefits of the proposed approach on some real world datasets.">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2020-11-22</td>
           <td>Biophysical Reviews</td>
-          <td>35</td>
+          <td>36</td>
           <td>5</td>
         </tr>
     
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2020-07-02</td>
           <td>ArXiv</td>
-          <td>129</td>
+          <td>136</td>
           <td>23</td>
         </tr>
     
@@ -138,6 +138,18 @@ hide:
           <td>4</td>
         </tr>
     
+        <tr id="None">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/a9a3238808d20c5236a7c5f912c20e6dd729f88c" target='_blank'>Data-driven Reconstruction of Nonlinear Dynamics from Sparse Observation</a></td>
+          <td>
+            K. Yeo
+          </td>
+          <td>2019-06-10</td>
+          <td>J. Comput. Phys.</td>
+          <td>23</td>
+          <td>19</td>
+        </tr>
+    
         <tr id="In recent years, we have witnessed a significant shift toward ever-more complex and ever-larger-scale systems in the majority of the grand societal challenges tackled in applied sciences. The need to comprehend and predict the dynamics of complex systems have spurred developments in large-scale simulations and a multitude of methods across several disciplines. The goals of understanding and prediction in complex dynamical systems, however, have been hindered by high dimensionality, complexity and chaotic behaviours. Recent advances in data-driven techniques and machine-learning approaches have revolutionized how we model and analyse complex systems. The integration of these techniques with dynamical systems theory opens up opportunities to tackle previously unattainable challenges in modelling and prediction of dynamical systems. While data-driven prediction methods have made great strides in recent years, it is still necessary to develop new techniques to improve their applicability to a wider range of complex systems in science and engineering. This focus issue shares recent developments in the field of complex dynamical systems with emphasis on data-driven, data-assisted and artificial intelligence-based discovery of dynamical systems. This article is part of the theme issue 'Data-driven prediction in dynamical systems'.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/e5aeec169d2c4d5e8e730a3c382006ba983b8709" target='_blank'>Data-driven prediction in dynamical systems: recent developments</a></td>
@@ -146,22 +158,10 @@ hide:
           </td>
           <td>2022-06-20</td>
           <td>Philosophical transactions. Series A, Mathematical, physical, and engineering sciences</td>
-          <td>43</td>
+          <td>48</td>
           <td>33</td>
         </tr>
     
-        <tr id="This article discusses a newly developed online manifold learning method, subspace iteration using reduced models (SIRM), for the dimensionality reduction of dynamical systems. This method may be viewed as subspace iteration combined with a model reduction procedure. Specifically, starting with a test solution, the method solves a reduced model to obtain a more precise solution, and it repeats this process until sufficient accuracy is achieved. The reduced model is obtained by projecting the full model onto a subspace that is spanned by the dominant modes of an extended data ensemble. The extended data ensemble in this article contains not only the state vectors of some snapshots of the approximate solution from the previous iteration but also the associated tangent vectors. Therefore, the proposed manifold learning method takes advantage of the information of the original dynamical system to reduce the dynamics. Moreover, the learning procedure is computed in the online stage, as opposed to being compute...">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/418ebf87dea83c922735e9083fe533efc8a04b7f" target='_blank'>An Online Manifold Learning Approach for Model Reduction of Dynamical Systems</a></td>
-          <td>
-            Liqian Peng, K. Mohseni
-          </td>
-          <td>2012-10-10</td>
-          <td>SIAM J. Numer. Anal.</td>
-          <td>22</td>
-          <td>40</td>
-        </tr>
-    
   </tbody>
   <tfoot>
     <tr>
diff --git a/docs/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220.md b/docs/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220.md
index 7500f784..c1aa11ce 100644
--- a/docs/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220.md
+++ b/docs/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:05 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:50 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2019-06-25</td>
           <td>IEEE Access</td>
-          <td>98</td>
-          <td>65</td>
+          <td>103</td>
+          <td>67</td>
         </tr>
     
         <tr id="Energy conservation is a basic physics principle, the breakdown of which often implies new physics. This paper presents a method for data-driven "new physics" discovery. Specifically, given a trajectory governed by unknown forces, our neural new-physics detector (NNPhD) aims to detect new physics by decomposing the force field into conservative and nonconservative components, which are represented by a Lagrangian neural network (LNN) and an unconstrained neural network, respectively, trained to minimize the force recovery error plus a constant λ times the magnitude of the predicted nonconservative force. We show that a phase transition occurs at λ=1, universally for arbitrary forces. We demonstrate that NNPhD successfully discovers new physics in toy numerical experiments, rediscovering friction (1493) from a damped double pendulum, Neptune from Uranus' orbit (1846), and gravitational waves (2017) from an inspiraling orbit. We also show how NNPhD coupled with an integrator outperforms both an LNN and an unconstrained neural network for predicting the future of a damped double pendulum.">
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2021-05-31</td>
           <td>Physical review. E</td>
-          <td>14</td>
-          <td>82</td>
+          <td>15</td>
+          <td>83</td>
         </tr>
     
         <tr id="Invariants and conservation laws convey critical information about the underlying dynamics of a system, yet it is generally infeasible to find them from large-scale data without any prior knowledge or human insight. We propose ConservNet to achieve this goal, a neural network that spontaneously discovers a conserved quantity from grouped data where the members of each group share invariants, similar to a general experimental setting where trajectories from different trials are observed. As a neural network trained with a novel and intuitive loss function called noise-variance loss, ConservNet learns the hidden invariants in each group of multi-dimensional observables in a data-driven, end-to-end manner. Our model successfully discovers underlying invariants from the simulated systems having invariants as well as a real-world double pendulum trajectory. Since the model is robust to various noises and data conditions compared to baseline, our approach is directly applicable to experimental data for discovering hidden conservation laws and further, general relationships between variables.">
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2023-05-31</td>
           <td>2023 American Control Conference (ACC)</td>
-          <td>20</td>
+          <td>22</td>
           <td>38</td>
         </tr>
     
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2022-02-17</td>
           <td>Communications Physics</td>
-          <td>23</td>
-          <td>30</td>
+          <td>25</td>
+          <td>31</td>
         </tr>
     
         <tr id="We present AI Poincaré, a machine learning algorithm for autodiscovering conserved quantities using trajectory data from unknown dynamical systems. We test it on five Hamiltonian systems, including the gravitational three-body problem, and find that it discovers not only all exactly conserved quantities, but also periodic orbits, phase transitions, and breakdown timescales for approximate conservation laws.">
@@ -134,7 +134,7 @@ hide:
           </td>
           <td>2021-05-06</td>
           <td>Physical review letters</td>
-          <td>87</td>
+          <td>93</td>
           <td>82</td>
         </tr>
     
@@ -146,7 +146,7 @@ hide:
           </td>
           <td>2023-09-01</td>
           <td>Nature Computational Science</td>
-          <td>0</td>
+          <td>1</td>
           <td>17</td>
         </tr>
     
@@ -158,8 +158,8 @@ hide:
           </td>
           <td>2021-07-06</td>
           <td>ArXiv</td>
-          <td>23</td>
-          <td>27</td>
+          <td>25</td>
+          <td>28</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/3c9961153493370500020c81527b3548c96f81e0.md b/docs/recommendations/3c9961153493370500020c81527b3548c96f81e0.md
index 3e11f2f1..10eed999 100644
--- a/docs/recommendations/3c9961153493370500020c81527b3548c96f81e0.md
+++ b/docs/recommendations/3c9961153493370500020c81527b3548c96f81e0.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:55 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:42 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
     
         <tr id="None">
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>250</td>
-          <td>12</td>
+          <td>289</td>
+          <td>13</td>
         </tr>
     
         <tr id="Harnessing data to discover the underlying governing laws or equations that describe the behavior of complex physical systems can significantly advance our modeling, simulation and understanding of such systems in various science and engineering disciplines. Recent advances in sparse identification show encouraging success in distilling closed-form governing equations from data for a wide range of nonlinear dynamical systems. However, the fundamental bottleneck of this approach lies in the robustness and scalability with respect to data scarcity and noise. This work introduces a novel physics-informed deep learning framework to discover governing partial differential equations (PDEs) from scarce and noisy data for nonlinear spatiotemporal systems. In particular, this approach seamlessly integrates the strengths of deep neural networks for rich representation learning, automatic differentiation and sparse regression to approximate the solution of system variables, compute essential derivatives, as well as identify the key derivative terms and parameters that form the structure and explicit expression of the PDEs. The efficacy and robustness of this method are demonstrated on discovering a variety of PDE systems with different levels of data scarcity and noise. The resulting computational framework shows the potential for closed-form model discovery in practical applications where large and accurate datasets are intractable to capture.">
@@ -75,7 +75,7 @@ hide:
           <td>2020-05-05</td>
           <td>ArXiv</td>
           <td>19</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
     
         <tr id="None">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2020-11-22</td>
           <td>Biophysical Reviews</td>
-          <td>35</td>
+          <td>36</td>
           <td>5</td>
         </tr>
     
@@ -98,8 +98,8 @@ hide:
           </td>
           <td>2022-11-19</td>
           <td>Proceedings of the Royal Society A</td>
-          <td>12</td>
-          <td>31</td>
+          <td>13</td>
+          <td>32</td>
         </tr>
     
         <tr id="Identifying governing equations for a dynamical system is a topic of critical interest across an array of disciplines, from mathematics to engineering to biology. Machine learning-specifically deep learning-techniques have shown their capabilities in approximating dynamics from data, but a shortcoming of traditional deep learning is that there is little insight into the underlying mapping beyond its numerical output for a given input. This limits their utility in analysis beyond simple prediction. Simultaneously, a number of strategies exist which identify models based on a fixed dictionary of basis functions, but most either require some intuition or insight about the system, or are susceptible to overfitting or a lack of parsimony. Here, we present a novel approach that combines the flexibility and accuracy of deep learning approaches with the utility of symbolic solutions: a deep neural network that generates a symbolic expression for the governing equations. We first describe the architecture for our model and then show the accuracy of our algorithm across a range of classical dynamical systems.">
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2023-08-01</td>
           <td>Chaos</td>
-          <td>2</td>
+          <td>4</td>
           <td>37</td>
         </tr>
     
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
     
@@ -134,8 +134,8 @@ hide:
           </td>
           <td>2021-07-22</td>
           <td>Communications Physics</td>
-          <td>17</td>
-          <td>94</td>
+          <td>20</td>
+          <td>95</td>
         </tr>
     
         <tr id="Estimating governing equations from observed time-series data is crucial for understanding dynamical systems. From the perspective of system comprehension, the demand for accurate estimation and interpretable results has been particularly emphasized. Herein, we propose a novel data-driven method for estimating the governing equations of dynamical systems based on machine learning with high accuracy and interpretability. The proposed method enhances the estimation accuracy for dynamical systems using sparse modeling by incorporating physical constraints derived from Hamiltonian mechanics. Unlike conventional approaches used for estimating governing equations for dynamical systems, we employ a sparse representation of Hamiltonian, allowing for the estimation. Using noisy observational data, the proposed method demonstrates a capability to achieve accurate parameter estimation and extraction of essential nonlinear terms. In addition, it is shown that estimations based on energy conservation principles exhibit superior accuracy in long-term predictions. These results collectively indicate that the proposed method accurately estimates dynamical systems while maintaining interpretability.">
@@ -159,7 +159,7 @@ hide:
           <td>2021-06-09</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2.md b/docs/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2.md
index 9b2d99cf..c8ce8480 100644
--- a/docs/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2.md
+++ b/docs/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:56 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:42 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2022-11-20</td>
           <td>Physical review. E</td>
-          <td>9</td>
+          <td>10</td>
           <td>51</td>
         </tr>
     
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2022-01-13</td>
           <td>Proceedings of the Royal Society A</td>
-          <td>70</td>
-          <td>65</td>
+          <td>75</td>
+          <td>67</td>
         </tr>
     
         <tr id="Making accurate forecasts for a complex system is a challenge in various practical applications. The major difficulty in solving such a problem concerns nonlinear spatiotemporal dynamics with time-varying characteristics. Takens' delay embedding theory provides a way to transform high-dimensional spatial information into temporal information. In this work, by combining delay embedding theory and deep learning techniques, we propose a novel framework, delay-embedding-based forecast Machine (DEFM), to predict the future values of a target variable in a self-supervised and multistep-ahead manner based on high-dimensional observations. With a three-module spatiotemporal architecture, the DEFM leverages deep neural networks to effectively extract both the spatially and temporally associated information from the observed time series even with time-varying parameters or additive noise. The DEFM can accurately predict future information by transforming spatiotemporal information to the delay embeddings of a target variable. The efficacy and precision of the DEFM are substantiated through applications in three spatiotemporally chaotic systems: a 90-dimensional (90D) coupled Lorenz system, the Lorenz 96 system, and the Kuramoto-Sivashinsky equation with inhomogeneity. Additionally, the performance of the DEFM is evaluated on six real-world datasets spanning various fields. Comparative experiments with five prediction methods illustrate the superiority and robustness of the DEFM and show the great potential of the DEFM in temporal information mining and forecasting.">
@@ -87,7 +87,7 @@ hide:
           <td>2020-05-16</td>
           <td>Chaos</td>
           <td>1</td>
-          <td>80</td>
+          <td>83</td>
         </tr>
     
         <tr id="Making accurate forecasts for a complex system is a challenge in various practical applications. The major difficulty in solving such a problem concerns nonlinear spatiotemporal dynamics with time-varying characteristics. Takens' delay embedding theory provides a way to transform high-dimensional spatial information into temporal information. In this work, by combining delay embedding theory and deep learning techniques, we propose a novel framework, delay-embedding-based forecast Machine (DEFM), to predict the future values of a target variable in a self-supervised and multistep-ahead manner based on high-dimensional observations. With a three-module spatiotemporal architecture, the DEFM leverages deep neural networks to effectively extract both the spatially and temporally associated information from the observed time series even with time-varying parameters or additive noise. The DEFM can accurately predict future information by transforming spatiotemporal information to the delay embeddings of a target variable. The efficacy and precision of the DEFM are substantiated through applications in three spatiotemporally chaotic systems: a 90-dimensional (90D) coupled Lorenz system, the Lorenz 96 system, and the Kuramoto-Sivashinsky equation with inhomogeneity. Additionally, the performance of the DEFM is evaluated on six real-world datasets spanning various fields. Comparative experiments with five prediction methods illustrate the superiority and robustness of the DEFM and show the great potential of the DEFM in temporal information mining and forecasting.">
@@ -99,7 +99,7 @@ hide:
           <td>2020-05-16</td>
           <td>Chaos</td>
           <td>1</td>
-          <td>80</td>
+          <td>83</td>
         </tr>
     
         <tr id="Many complex phenomena, from weather systems to heartbeat rhythm patterns, are effectively modeled as low-dimensional dynamical systems. Such systems may behave chaotically under certain conditions, and so the ability to detect chaos based on empirical measurement is an important step in characterizing and predicting these processes. Classifying a system as chaotic usually requires estimating its largest Lyapunov exponent, which quantifies the average rate of convergence or divergence of initially close trajectories in state space, and for which a positive value is generally accepted as an operational definition of chaos. Estimating the largest Lyapunov exponent from observations of a process is especially challenging in systems affected by dynamical noise, which is the case for many models of real-world processes, in particular models of biological systems. We describe a novel method for estimating the largest Lyapunov exponent from data, based on training Deep Learning models on synthetically generated trajectories, and demonstrate that this method yields accurate and noise-robust predictions given relatively short inputs and across a range of different dynamical systems. Our method is unique in that it can analyze tree-shaped data, a ubiquitous topology in biological settings, and specifically in dynamics over lineages of cells or organisms. We also characterize the types of input information extracted by our models for their predictions, allowing for a deeper understanding into the different ways by which chaos can be analyzed in different topologies.">
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2021-06-01</td>
           <td>Science Advances</td>
-          <td>44</td>
-          <td>49</td>
+          <td>50</td>
+          <td>50</td>
         </tr>
     
         <tr id="We propose a universal method for data-driven modeling of complex nonlinear dynamics from time-resolved snapshot data without prior knowledge. Complex nonlinear dynamics govern many fields of science and engineering. Data-driven dynamic modeling often assumes a low-dimensional subspace or manifold for the state. We liberate ourselves from this assumption by proposing cluster-based network modeling (CNM) bridging machine learning, network science, and statistical physics. CNM only assumes smoothness of the dynamics in the state space, robustly describes short- and long-term behavior and is fully automatable as it does not rely on application-specific knowledge. CNM is demonstrated for the Lorenz attractor, ECG heartbeat signals, Kolmogorov flow, and a high-dimensional actuated turbulent boundary layer. Even the notoriously difficult modeling benchmark of rare events in the Kolmogorov flow is solved. This automatable universal data-driven representation of complex nonlinear dynamics complements and expands network connectivity science and promises new fast-track avenues to understand, estimate, predict and control complex systems in all scientific fields.">
@@ -135,7 +135,7 @@ hide:
           <td>2020-10-30</td>
           <td>arXiv: Data Analysis, Statistics and Probability</td>
           <td>1</td>
-          <td>49</td>
+          <td>50</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba.md b/docs/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba.md
index efd5e825..3286942f 100644
--- a/docs/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba.md
+++ b/docs/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:09 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:08 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2023-06-26</td>
           <td>DBLP</td>
-          <td>9</td>
-          <td>16</td>
+          <td>12</td>
+          <td>17</td>
         </tr>
     
         <tr id="Irregularly-sampled time series (ITS) are native to high-impact domains like healthcare, where measurements are collected over time at uneven intervals. However, for many classification problems, only small portions of long time series are often relevant to the class label. In this case, existing ITS models often fail to classify long series since they rely on careful imputation, which easily over- or under-samples the relevant regions. Using this insight, we then propose CAT, a model that classifies multivariate ITS by explicitly seeking highly-relevant portions of an input series' timeline. CAT achieves this by integrating three components: (1) A Moment Network learns to seek relevant moments in an ITS's continuous timeline using reinforcement learning. (2) A Receptor Network models the temporal dynamics of both observations and their timing localized around predicted moments. (3) A recurrent Transition Model models the sequence of transitions between these moments, cultivating a representation with which the series is classified. Using synthetic and real data, we find that CAT outperforms ten state-of-the-art methods by finding short signals in long irregular time series.">
@@ -63,7 +63,7 @@ hide:
           <td>2023-02-08</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>38</td>
+          <td>39</td>
         </tr>
     
         <tr id="Prevalent in many real-world settings such as healthcare, irregular time series are challenging to formulate predictions from. It is difficult to infer the value of a feature at any given time when observations are sporadic, as it could take on a range of values depending on when it was last observed. To characterize this uncertainty we present EDICT, a strategy that learns an evidential distribution over irregular time series in continuous time. This distribution enables well-calibrated and flexible inference of partially observed features at any time of interest, while expanding uncertainty temporally for sparse, irregular observations. We demonstrate that EDICT attains competitive performance on challenging time series classification tasks and enabling uncertainty-guided inference when encountering noisy data.">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2023-03-01</td>
           <td>ArXiv</td>
-          <td>12</td>
+          <td>17</td>
           <td>65</td>
         </tr>
     
@@ -98,10 +98,22 @@ hide:
           </td>
           <td>2023-03-01</td>
           <td>ArXiv</td>
-          <td>12</td>
+          <td>17</td>
           <td>65</td>
         </tr>
     
+        <tr id="Irregular time series, where data points are recorded at uneven intervals, are prevalent in healthcare settings, such as emergency wards where vital signs and laboratory results are captured at varying times. This variability, which reflects critical fluctuations in patient health, is essential for informed clinical decision-making. Existing self-supervised learning research on irregular time series often relies on generic pretext tasks like forecasting, which may not fully utilise the signal provided by irregular time series. There is a significant need for specialised pretext tasks designed for the characteristics of irregular time series to enhance model performance and robustness, especially in scenarios with limited data availability. This paper proposes a novel pretraining framework, EMIT, an event-based masking for irregular time series. EMIT focuses on masking-based reconstruction in the latent space, selecting masking points based on the rate of change in the data. This method preserves the natural variability and timing of measurements while enhancing the model's ability to process irregular intervals without losing essential information. Extensive experiments on the MIMIC-III and PhysioNet Challenge datasets demonstrate the superior performance of our event-based masking strategy. The code has been released at https://github.com/hrishi-ds/EMIT.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/5e13c5354481232bed4c0fb1e4b6a37d07f32e52" target='_blank'>EMIT- Event-Based Masked Auto Encoding for Irregular Time Series</a></td>
+          <td>
+            Hrishikesh Patel, Ruihong Qiu, Adam Irwin, S. Sadiq, Sen Wang
+          </td>
+          <td>2024-09-25</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>2</td>
+        </tr>
+    
         <tr id="To analyze multivariate time series, most previous methods assume regular subsampling of time series, where the interval between adjacent measurements and the number of samples remain unchanged. Practically, data collection systems could produce irregularly sampled time series due to sensor failures and interventions. However, existing methods designed for regularly sampled multivariate time series cannot directly handle irregularity owing to misalignment along both temporal and variate dimensions. To fill this gap, we propose Compatible Transformer (CoFormer), a transformer-based encoder to achieve comprehensive temporal-interaction feature learning for each individual sample in irregular multivariate time series. In CoFormer, we view each sample as a unique variate-time point and leverage intra-variate/inter-variate attentions to learn sample-wise temporal/interaction features based on intra-variate/inter-variate neighbors. With CoFormer as the core, we can analyze irregularly sampled multivariate time series for many downstream tasks, including classification and prediction. We conduct extensive experiments on 3 real-world datasets and validate that the proposed CoFormer significantly and consistently outperforms existing methods. Code will be avilable at https://github.com/MediaBrain-SJTU/CoFormer.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/f6f708484a85b2b96d6f6764724b950cab167bb7" target='_blank'>Compatible Transformer for Irregularly Sampled Multivariate Time Series</a></td>
@@ -121,9 +133,9 @@ hide:
             Steven Cheng-Xian Li, Benjamin M Marlin
           </td>
           <td>2020-07-12</td>
-          <td>ArXiv, DBLP, MAG</td>
-          <td>48</td>
-          <td>33</td>
+          <td>ArXiv, MAG, DBLP</td>
+          <td>54</td>
+          <td>34</td>
         </tr>
     
         <tr id="We consider a sequence of related multivariate time series learning tasks, such as predicting failures for different instances of a machine from time series of multi-sensor data, or activity recognition tasks over different individuals from multiple wearable sensors. We focus on two under-explored practical challenges arising in such settings: (i) Each task may have a different subset of sensors, i.e., providing different partial observations of the underlying ‘system’. This restriction can be due to different manufacturers in the former case, and people wearing more or less measurement devices in the latter (ii) We are not allowed to store or re-access data from a task once it has been observed at the task level. This may be due to privacy considerations in the case of people, or legal restrictions placed by machine owners. Nevertheless, we would like to (a) improve performance on subsequent tasks using experience from completed tasks as well as (b) continue to perform better on past tasks, e.g., update the model and improve predictions on even the first machine after learning from subsequently observed ones. We note that existing continual learning methods do not take into account variability in input dimensions arising due to different subsets of sensors being available across tasks, and struggle to adapt to such variable input dimensions (VID) tasks. In this work, we address this shortcoming of existing methods. To this end, we learn task-specific generative models and classifiers, and use these to augment data for target tasks. Since the input dimensions across tasks vary, we propose a novel conditioning module based on graph neural networks to aid a standard recurrent neural network. We evaluate the efficacy of the proposed approach on three publicly available datasets corresponding to two activity recognition tasks (classification) and one prognostics task (regression). We demonstrate that it is possible to significantly enhance the performance on future and previous tasks while learning continuously from VID tasks without storing data.">
@@ -134,20 +146,8 @@ hide:
           </td>
           <td>2021-12-01</td>
           <td>2021 IEEE International Conference on Data Mining (ICDM)</td>
-          <td>16</td>
-          <td>27</td>
-        </tr>
-    
-        <tr id="Multivariate time series (MTS) data are becoming increasingly ubiquitous in diverse domains, e.g., IoT systems, health informatics, and 5G networks. To obtain an effective representation of MTS data, it is not only essential to consider unpredictable dynamics and highly variable lengths of these data but also important to address the irregularities in the sampling rates of MTS. Existing parametric approaches rely on manual hyperparameter tuning and may cost a huge amount of labor effort. Therefore, it is desirable to learn the representation automatically and efficiently. To this end, we propose an autonomous representation learning approach for multivariate time series (TimeAutoML) with irregular sampling rates and variable lengths. As opposed to previous works, we first present a representation learning pipeline in which the configuration and hyperparameter optimization are fully automatic and can be tailored for various tasks, e.g., anomaly detection, clustering, etc. Next, a negative sample generation approach and an auxiliary classification task are developed and integrated within TimeAutoML to enhance its representation capability. Extensive empirical studies on real-world datasets demonstrate that the proposed TimeAutoML outperforms competing approaches on various tasks by a large margin. In fact, it achieves the best anomaly detection performance among all comparison algorithms on 78 out of all 85 UCR datasets, acquiring up to 20% performance improvement in terms of AUC score.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/60773c01dd3c76be0e1d3bc429cf474e5f9d8519" target='_blank'>TimeAutoML: Autonomous Representation Learning for Multivariate Irregularly Sampled Time Series</a></td>
-          <td>
-            Yang Jiao, Kai Yang, Shaoyu Dou, Pan Luo, Sijia Liu, Dongjin Song
-          </td>
-          <td>2020-10-04</td>
-          <td>ArXiv</td>
-          <td>6</td>
-          <td>22</td>
+          <td>17</td>
+          <td>28</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01.md b/docs/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01.md
index bbf06685..5b5d1197 100644
--- a/docs/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01.md
+++ b/docs/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:03 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:46 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2022-09-04</td>
           <td>Scientific Reports</td>
-          <td>2</td>
-          <td>23</td>
+          <td>3</td>
+          <td>24</td>
         </tr>
     
         <tr id="Modern societies have an abundance of data yet good system models are rare. Unfortunately, many of the current system identification and machine learning techniques fail to generalize outside of the training set, producing models that violate basic physical laws. This work proposes a novel method for the Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI). SINDy-SI is an iterative method that uses Sum-of-Squares (SOS) programming to learn optimally fitted models while guaranteeing that the learned model satisfies side information, such as symmetries and physical laws. Guided by the principle of Occam's razor, that the simplest or most regularized best fitted model is typically the superior choice, during each iteration SINDy-SI prunes the basis functions associated with small coefficients, yielding a sparse dynamical model upon termination. Through several numerical experiments we will show how the combination of side information constraints and sparse polynomial representation cultivates dynamical models that obey known physical laws while displaying impressive generalized performance beyond the training set.">
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>111</td>
-          <td>65</td>
+          <td>119</td>
+          <td>67</td>
         </tr>
     
         <tr id="Data-driven modeling frameworks that adopt sparse regression techniques, such as sparse identification of nonlinear dynamics (SINDy) and its modifications, are developed to resolve difficulties in extracting underlying dynamics from experimental data. In contrast to neural-network-based methods, these methods are designed to obtain white-box analytical models. In this work, we incorporate the concept of SINDy and knowledge in the field of classical mechanics to identify interpretable and sparse expressions of total energy and the Lagrangian that shelters the hidden dynamics. Moreover, our method (hereafter referred as Lagrangian-SINDy) is developed to use knowledge of simple systems that form the system being analyzed to ensure the likelihood of correct results and to improve the learning pace. Lagrangian-SINDy is highly accurate in discovering interpretable dynamics via energy-related physical quantities. Its performance is validated with three popular multi-DOF nonlinear dynamical systems, namely the spherical pendulum, double pendulum and cart-pendulum system. Comparisons with other SINDy-based methods are made and Lagrangian-SINDy is found to provide the most compact analytical models.">
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2020-01-31</td>
           <td>IEEE Robotics and Automation Letters</td>
-          <td>23</td>
-          <td>23</td>
+          <td>24</td>
+          <td>24</td>
         </tr>
     
         <tr id="None">
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
     
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2022-11-10</td>
           <td>SSRN Electronic Journal</td>
-          <td>10</td>
+          <td>14</td>
           <td>33</td>
         </tr>
     
@@ -118,12 +118,24 @@ hide:
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/7a3166991a8c4161d83c4cdb27033c02f3135cc0" target='_blank'>A Robust SINDy Approach by Combining Neural Networks and an Integral Form</a></td>
           <td>
-            A. Forootani, P. Goyal, P. Benner
+            Ali Forootani, P. Goyal, P. Benner
           </td>
           <td>2023-09-13</td>
           <td>ArXiv</td>
           <td>2</td>
-          <td>14</td>
+          <td>15</td>
+        </tr>
+    
+        <tr id="Identifying dynamical systems characterized by nonlinear parameters presents significant challenges in deriving mathematical models that enhance understanding of physics. Traditional methods, such as Sparse Identification of Nonlinear Dynamics (SINDy) and symbolic regression, can extract governing equations from observational data; however, they also come with distinct advantages and disadvantages. This paper introduces a novel method within the SINDy framework, termed ADAM-SINDy, which synthesizes the strengths of established approaches by employing the ADAM optimization algorithm. This facilitates the simultaneous optimization of nonlinear parameters and coefficients associated with nonlinear candidate functions, enabling precise parameter estimation without requiring prior knowledge of nonlinear characteristics such as trigonometric frequencies, exponential bandwidths, or polynomial exponents, thereby addressing a key limitation of SINDy. Through an integrated global optimization, ADAM-SINDy dynamically adjusts all unknown variables in response to data, resulting in an adaptive identification procedure that reduces the sensitivity to the library of candidate functions. The performance of the ADAM-SINDy methodology is demonstrated across a spectrum of dynamical systems, including benchmark coupled nonlinear ordinary differential equations such as oscillators, chaotic fluid flows, reaction kinetics, pharmacokinetics, as well as nonlinear partial differential equations (wildfire transport). The results demonstrate significant improvements in identifying parameterized dynamical systems and underscore the importance of concurrently optimizing all parameters, particularly those characterized by nonlinear parameters. These findings highlight the potential of ADAM-SINDy to extend the applicability of the SINDy framework in addressing more complex challenges in dynamical system identification.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/0c327ed253a2c7b658940860bcf0a9e077d634b6" target='_blank'>ADAM-SINDy: An Efficient Optimization Framework for Parameterized Nonlinear Dynamical System Identification</a></td>
+          <td>
+            Siva Viknesh, Younes Tatari, Amirhossein Arzani
+          </td>
+          <td>2024-10-21</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>24</td>
         </tr>
     
         <tr id="Symbolic Regression (SR) is a widely studied field of research that aims to infer symbolic expressions from data. A popular approach for SR is the Sparse Identification of Nonlinear Dynamical Systems (\sindy) framework, which uses sparse regression to identify governing equations from data. This study introduces an enhanced method, Nested SINDy, that aims to increase the expressivity of the SINDy approach thanks to a nested structure. Indeed, traditional symbolic regression and system identification methods often fail with complex systems that cannot be easily described analytically. Nested SINDy builds on the SINDy framework by introducing additional layers before and after the core SINDy layer. This allows the method to identify symbolic representations for a wider range of systems, including those with compositions and products of functions. We demonstrate the ability of the Nested SINDy approach to accurately find symbolic expressions for simple systems, such as basic trigonometric functions, and sparse (false but accurate) analytical representations for more complex systems. Our results highlight Nested SINDy's potential as a tool for symbolic regression, surpassing the traditional SINDy approach in terms of expressivity. However, we also note the challenges in the optimization process for Nested SINDy and suggest future research directions, including the designing of a more robust methodology for the optimization process. This study proves that Nested SINDy can effectively discover symbolic representations of dynamical systems from data, offering new opportunities for understanding complex systems through data-driven methods.">
@@ -147,19 +159,7 @@ hide:
           <td>2024-03-01</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>1</td>
-        </tr>
-    
-        <tr id="None">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/7be51140c9c66207fc1baa0ca92a691726cbd900" target='_blank'>Weak SINDy For Partial Differential Equations</a></td>
-          <td>
-            D. Messenger, D. Bortz
-          </td>
-          <td>2020-07-06</td>
-          <td>Journal of computational physics</td>
-          <td>126</td>
-          <td>20</td>
+          <td>2</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/4a111f7a3b56d0468f13104999844885157ef17d.md b/docs/recommendations/4a111f7a3b56d0468f13104999844885157ef17d.md
index 8bf74e8b..7002d394 100644
--- a/docs/recommendations/4a111f7a3b56d0468f13104999844885157ef17d.md
+++ b/docs/recommendations/4a111f7a3b56d0468f13104999844885157ef17d.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:24 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:17 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -42,15 +42,51 @@ hide:
   </thead>
   <tbody>
     
+        <tr id="Time series foundation models excel in zero-shot forecasting, handling diverse tasks without explicit training. However, the advancement of these models has been hindered by the lack of comprehensive benchmarks. To address this gap, we introduce the General Time Series Forecasting Model Evaluation, GIFT-Eval, a pioneering benchmark aimed at promoting evaluation across diverse datasets. GIFT-Eval encompasses 28 datasets over 144,000 time series and 177 million data points, spanning seven domains, 10 frequencies, multivariate inputs, and prediction lengths ranging from short to long-term forecasts. To facilitate the effective pretraining and evaluation of foundation models, we also provide a non-leaking pretraining dataset containing approximately 230 billion data points. Additionally, we provide a comprehensive analysis of 17 baselines, which includes statistical models, deep learning models, and foundation models. We discuss each model in the context of various benchmark characteristics and offer a qualitative analysis that spans both deep learning and foundation models. We believe the insights from this analysis, along with access to this new standard zero-shot time series forecasting benchmark, will guide future developments in time series foundation models. The codebase, datasets, and a leaderboard showing all the results in detail will be available soon.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/8bf78baccb9e56ffef46080b9b34752f23d1911c" target='_blank'>GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation</a></td>
+          <td>
+            Taha İbrahim Aksu, Gerald Woo, Juncheng Liu, Xu Liu, Chenghao Liu, Silvio Savarese, Caiming Xiong, Doyen Sahoo
+          </td>
+          <td>2024-10-14</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>25</td>
+        </tr>
+    
+        <tr id="Time Series Forecasting (TSF) is key functionality in numerous fields, including in finance, weather services, and energy management. While TSF methods are emerging these days, many of them require domain-specific data collection and model training and struggle with poor generalization performance on new domains. Foundation models aim to overcome this limitation. Pre-trained on large-scale language or time series data, they exhibit promising inferencing capabilities in new or unseen data. This has spurred a surge in new TSF foundation models. We propose a new benchmark, FoundTS, to enable thorough and fair evaluation and comparison of such models. FoundTS covers a variety of TSF foundation models, including those based on large language models and those pretrained on time series. Next, FoundTS supports different forecasting strategies, including zero-shot, few-shot, and full-shot, thereby facilitating more thorough evaluations. Finally, FoundTS offers a pipeline that standardizes evaluation processes such as dataset splitting, loading, normalization, and few-shot sampling, thereby facilitating fair evaluations. Building on this, we report on an extensive evaluation of TSF foundation models on a broad range of datasets from diverse domains and with different statistical characteristics. Specifically, we identify pros and cons and inherent limitations of existing foundation models, and we identify directions for future model design. We make our code and datasets available at https://anonymous.4open.science/r/FoundTS-C2B0.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/c245a7d561855d7bef6c758e45e772129b52dacb" target='_blank'>FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting</a></td>
+          <td>
+            Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang
+          </td>
+          <td>2024-10-15</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>27</td>
+        </tr>
+    
+        <tr id="Deep learning for time series forecasting has seen significant advancements over the past decades. However, despite the success of large-scale pre-training in language and vision domains, pre-trained time series models remain limited in scale and operate at a high cost, hindering the development of larger capable forecasting models in real-world applications. In response, we introduce Time-MoE, a scalable and unified architecture designed to pre-train larger, more capable forecasting foundation models while reducing inference costs. By leveraging a sparse mixture-of-experts (MoE) design, Time-MoE enhances computational efficiency by activating only a subset of networks for each prediction, reducing computational load while maintaining high model capacity. This allows Time-MoE to scale effectively without a corresponding increase in inference costs. Time-MoE comprises a family of decoder-only transformer models that operate in an auto-regressive manner and support flexible forecasting horizons with varying input context lengths. We pre-trained these models on our newly introduced large-scale data Time-300B, which spans over 9 domains and encompassing over 300 billion time points. For the first time, we scaled a time series foundation model up to 2.4 billion parameters, achieving significantly improved forecasting precision. Our results validate the applicability of scaling laws for training tokens and model size in the context of time series forecasting. Compared to dense models with the same number of activated parameters or equivalent computation budgets, our models consistently outperform them by large margin. These advancements position Time-MoE as a state-of-the-art solution for tackling real-world time series forecasting challenges with superior capability, efficiency, and flexibility.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/2cb3044ef42c7ee022a988864028b80ce977072c" target='_blank'>Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts</a></td>
+          <td>
+            X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin
+          </td>
+          <td>2024-09-24</td>
+          <td>ArXiv</td>
+          <td>2</td>
+          <td>9</td>
+        </tr>
+    
         <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4fb78450650894091b0a55b5504a8b0a6f3dec37" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
           <td>
             Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>28</td>
+          <td>69</td>
           <td>18</td>
         </tr>
     
@@ -62,8 +98,8 @@ hide:
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>4</td>
-          <td>66</td>
+          <td>21</td>
+          <td>67</td>
         </tr>
     
         <tr id="We propose an efficient design of Transformer-based models for multivariate time series forecasting and self-supervised representation learning. It is based on two key components: (i) segmentation of time series into subseries-level patches which are served as input tokens to Transformer; (ii) channel-independence where each channel contains a single univariate time series that shares the same embedding and Transformer weights across all the series. Patching design naturally has three-fold benefit: local semantic information is retained in the embedding; computation and memory usage of the attention maps are quadratically reduced given the same look-back window; and the model can attend longer history. Our channel-independent patch time series Transformer (PatchTST) can improve the long-term forecasting accuracy significantly when compared with that of SOTA Transformer-based models. We also apply our model to self-supervised pre-training tasks and attain excellent fine-tuning performance, which outperforms supervised training on large datasets. Transferring of masked pre-trained representation on one dataset to others also produces SOTA forecasting accuracy. Code is available at: https://github.com/yuqinie98/PatchTST.">
@@ -74,56 +110,32 @@ hide:
           </td>
           <td>2022-11-27</td>
           <td>ArXiv</td>
-          <td>534</td>
-          <td>34</td>
+          <td>724</td>
+          <td>35</td>
         </tr>
     
-        <tr id="Time series forecasting is a critical and challenging task in practical application. Recent advancements in pre-trained foundation models for time series forecasting have gained significant interest. However, current methods often overlook the multi-scale nature of time series, which is essential for accurate forecasting. To address this, we propose HiMTM, a hierarchical multi-scale masked time series modeling with self-distillation for long-term forecasting. HiMTM integrates four key components: (1) hierarchical multi-scale transformer (HMT) to capture temporal information at different scales; (2) decoupled encoder-decoder (DED) that directs the encoder towards feature extraction while the decoder focuses on pretext tasks; (3) hierarchical self-distillation (HSD) for multi-stage feature-level supervision signals during pre-training; and (4) cross-scale attention fine-tuning (CSA-FT) to capture dependencies between different scales for downstream tasks. These components collectively enhance multi-scale feature extraction in masked time series modeling, improving forecasting accuracy. Extensive experiments on seven mainstream datasets show that HiMTM surpasses state-of-the-art self-supervised and end-to-end learning methods by a considerable margin of 3.16-68.54\%. Additionally, HiMTM outperforms the latest robust self-supervised learning method, PatchTST, in cross-domain forecasting by a significant margin of 2.3\%. The effectiveness of HiMTM is further demonstrated through its application in natural gas demand forecasting.">
+        <tr id="Developing a foundation model for time series forecasting across diverse domains has attracted significant attention in recent years. Existing works typically assume regularly sampled, well-structured data, limiting their applicability to more generalized scenarios where time series often contain missing values, unequal sequence lengths, and irregular time intervals between measurements. To cover diverse domains and handle variable regularities, we propose FlexTSF, a universal time series forecasting model that possesses better generalization and natively support both regular and irregular time series. FlexTSF produces forecasts in an autoregressive manner and incorporates three novel designs: VT-Norm, a normalization strategy to ablate data domain barriers, IVP Patcher, a patching module to learn representations from flexibly structured time series, and LED attention, an attention mechanism to seamlessly integrate these two and propagate forecasts with awareness of domain and time information. Experiments on 12 datasets show that FlexTSF outperforms state-of-the-art forecasting models respectively designed for regular and irregular time series. Furthermore, after self-supervised pre-training, FlexTSF shows exceptional performance in both zero-shot and few-show settings for time series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/d945eae22457648e191742c37bcad88a114ce703" target='_blank'>HiMTM: Hierarchical Multi-Scale Masked Time Series Modeling with Self-Distillation for Long-Term Forecasting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/45521591fd8c46068f0a8712b43ce6cdaac1edea" target='_blank'>FlexTSF: A Universal Forecasting Model for Time Series with Variable Regularities</a></td>
           <td>
-            Shubao Zhao, Ming Jin, Zhaoxiang Hou, Che-Sheng Yang, Zengxiang Li, Qingsong Wen, Yi Wang
+            Jingge Xiao, Yile Chen, Gao Cong, Wolfgang Nejdl, Simon Gottschalk
           </td>
-          <td>2024-01-10</td>
+          <td>2024-10-30</td>
           <td>ArXiv</td>
           <td>0</td>
           <td>6</td>
         </tr>
     
-        <tr id="The Transformer is a highly successful deep learning model that has revolutionised the world of artificial neural networks, first in natural language processing and later in computer vision. This model is based on the attention mechanism and is able to capture complex semantic relationships between a variety of patterns present in the input data. Precisely because of these characteristics, the Transformer has recently been exploited for time series forecasting problems, assuming a natural adaptability to the domain of continuous numerical series. Despite the acclaimed results in the literature, some works have raised doubts about the robustness and effectiveness of this approach. In this paper, we further investigate the effectiveness of Transformer-based models applied to the domain of time series forecasting, demonstrate their limitations, and propose a set of alternative models that are better performing and significantly less complex. In particular, we empirically show how simplifying Transformer-based forecasting models almost always leads to an improvement, reaching state of the art performance. We also propose shallow models without the attention mechanism, which compete with the overall state of the art in long time series forecasting, and demonstrate their ability to accurately predict time series over extremely long windows. From a methodological perspective, we show how it is always necessary to use a simple baseline to verify the effectiveness of proposed models, and finally, we conclude the paper with a reflection on recent research paths and the opportunity to follow trends and hypes even where it may not be necessary.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/b98bc7d48357932537fc9ba6035e3b6bd1018ae6" target='_blank'>Two Steps Forward and One Behind: Rethinking Time Series Forecasting with Deep Learning</a></td>
-          <td>
-            Riccardo Ughi, Eugenio Lomurno, Matteo Matteucci
-          </td>
-          <td>2023-04-10</td>
-          <td>ArXiv, DBLP</td>
-          <td>1</td>
-          <td>6</td>
-        </tr>
-    
-        <tr id="Time series has been left behind in the era of pre-training and transfer learning. While research in the fields of natural language processing and computer vision are enjoying progressively larger datasets to train massive models, the most popular time series datasets consist of only tens of thousands of time steps, limiting our ability to study the effectiveness of pre-training and scaling. Recent studies have also cast doubt on the need for expressive models and scale. To alleviate these issues, we introduce three large-scale time series forecasting datasets from the cloud operations (CloudOps) domain, the largest having billions of observations, enabling further study into pre-training and scaling of time series models. We build the empirical groundwork for studying pre-training and scaling of time series models and pave the way for future research by identifying a promising candidate architecture. We show that it is a strong zero-shot baseline and benefits from further scaling, both in model and dataset size. Accompanying these datasets and results is a suite of comprehensive benchmark results comparing classical and deep learning baselines to our pre-trained method - achieving a 27% reduction in error on the largest dataset. Code and datasets can be found https://github.com/SalesforceAIResearch/pretrain-time-series-cloudops.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/dbd6463cfff26861f6f3ea8e8d19eff4e6331904" target='_blank'>Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain</a></td>
-          <td>
-            Gerald Woo, Chenghao Liu, Akshat Kumar, Doyen Sahoo
-          </td>
-          <td>2023-10-08</td>
-          <td>ArXiv</td>
-          <td>7</td>
-          <td>22</td>
-        </tr>
-    
-        <tr id="Time Series Forecasting plays a crucial role in various fields such as industrial equipment maintenance, meteorology, energy consumption, traffic flow and financial investment. However, despite their considerable advantages over traditional statistical approaches, current deep learning-based predictive models often exhibit a significant deviation between their forecasting outcomes and the ground truth. This discrepancy is largely due to an insufficient emphasis on extracting the sequence's latent information, particularly its global information within the frequency domain and the relationship between different variables. To address this issue, we propose a novel model Frequency-domain Attention In Two Horizons, which decomposes time series into trend and seasonal components using a multi-scale sequence adaptive decomposition and fusion architecture, and processes them separately. FAITH utilizes Frequency Channel feature Extraction Module and Frequency Temporal feature Extraction Module to capture inter-channel relationships and temporal global information in the sequence, significantly improving its ability to handle long-term dependencies and complex patterns. Furthermore, FAITH achieves theoretically linear complexity by modifying the time-frequency domain transformation method, effectively reducing computational costs. Extensive experiments on 6 benchmarks for long-term forecasting and 3 benchmarks for short-term forecasting demonstrate that FAITH outperforms existing models in many fields, such as electricity, weather and traffic, proving its effectiveness and superiority both in long-term and short-term time series forecasting tasks. Our codes and data are available at https://github.com/LRQ577/FAITH.">
+        <tr id="Time series forecasting is a critical and challenging task in practical application. Recent advancements in pre-trained foundation models for time series forecasting have gained significant interest. However, current methods often overlook the multi-scale nature of time series, which is essential for accurate forecasting. To address this, we propose HiMTM, a hierarchical multi-scale masked time series modeling with self-distillation for long-term forecasting. HiMTM integrates four key components: (1) hierarchical multi-scale transformer (HMT) to capture temporal information at different scales; (2) decoupled encoder-decoder (DED) that directs the encoder towards feature extraction while the decoder focuses on pretext tasks; (3) hierarchical self-distillation (HSD) for multi-stage feature-level supervision signals during pre-training; and (4) cross-scale attention fine-tuning (CSA-FT) to capture dependencies between different scales for downstream tasks. These components collectively enhance multi-scale feature extraction in masked time series modeling, improving forecasting accuracy. Extensive experiments on seven mainstream datasets show that HiMTM surpasses state-of-the-art self-supervised and end-to-end learning methods by a considerable margin of 3.16-68.54\%. Additionally, HiMTM outperforms the latest robust self-supervised learning method, PatchTST, in cross-domain forecasting by a significant margin of 2.3\%. The effectiveness of HiMTM is further demonstrated through its application in natural gas demand forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/ca89b6de0dbe31f523495727dc83f8ab5d2ae1c3" target='_blank'>FAITH: Frequency-domain Attention In Two Horizons for Time Series Forecasting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/d945eae22457648e191742c37bcad88a114ce703" target='_blank'>HiMTM: Hierarchical Multi-Scale Masked Time Series Modeling with Self-Distillation for Long-Term Forecasting</a></td>
           <td>
-            Ruiqi Li, Maowei Jiang, Kai Wang, Kaiduo Feng, Quangao Liu, Yue Sun, Xiufang Zhou
+            Shubao Zhao, Ming Jin, Zhaoxiang Hou, Che-Sheng Yang, Zengxiang Li, Qingsong Wen, Yi Wang
           </td>
-          <td>2024-05-22</td>
-          <td>ArXiv</td>
+          <td>2024-01-10</td>
+          <td>Proceedings of the 33rd ACM International Conference on Information and Knowledge Management</td>
           <td>0</td>
-          <td>2</td>
+          <td>7</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14.md b/docs/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14.md
index 72876976..a8c3d1db 100644
--- a/docs/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14.md
+++ b/docs/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:32 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:23 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -63,7 +63,7 @@ hide:
           <td>2024-08-28</td>
           <td>IEEE transactions on neural networks and learning systems</td>
           <td>0</td>
-          <td>14</td>
+          <td>15</td>
         </tr>
     
         <tr id="Unsupervised domain adaptation methods aim at generalizing well on unlabeled test data that may have a different (shifted) distribution from the training data. Such methods are typically developed on image data, and their application to time series data is less explored. Existing works on time series domain adaptation suffer from inconsistencies in evaluation schemes, datasets, and backbone neural network architectures. Moreover, labeled target data are often used for model selection, which violates the fundamental assumption of unsupervised domain adaptation. To address these issues, we develop a benchmarking evaluation suite (AdaTime) to systematically and fairly evaluate different domain adaptation methods on time series data. Specifically, we standardize the backbone neural network architectures and benchmarking datasets, while also exploring more realistic model selection approaches that can work with no labeled data or just a few labeled samples. Our evaluation includes adapting state-of-the-art visual domain adaptation methods to time series data as well as the recent methods specifically developed for time series data. We conduct extensive experiments to evaluate 11 state-of-the-art methods on five representative datasets spanning 50 cross-domain scenarios. Our results suggest that with careful selection of hyper-parameters, visual domain adaptation methods are competitive with methods proposed for time series domain adaptation. In addition, we find that hyper-parameters could be selected based on realistic model selection approaches. Our work unveils practical insights for applying domain adaptation methods on time series data and builds a solid foundation for future works in the field. The code is available at github.com/emadeldeen24/AdaTime.">
@@ -74,8 +74,32 @@ hide:
           </td>
           <td>2022-03-15</td>
           <td>ACM Transactions on Knowledge Discovery from Data</td>
-          <td>40</td>
-          <td>37</td>
+          <td>51</td>
+          <td>39</td>
+        </tr>
+    
+        <tr id="Unsupervised domain adaptation (UDA) of time series aims to teach models to identify consistent patterns across various temporal scenarios, disregarding domain-specific differences, which can maintain their predictive accuracy and effectively adapt to new domains. However, existing UDA methods struggle to adequately extract and align both global and local features in time series data. To address this issue, we propose the Local-Global Representation Alignment framework (LogoRA), which employs a two-branch encoder, comprising a multi-scale convolutional branch and a patching transformer branch. The encoder enables the extraction of both local and global representations from time series. A fusion module is then introduced to integrate these representations, enhancing domain-invariant feature alignment from multi-scale perspectives. To achieve effective alignment, LogoRA employs strategies like invariant feature learning on the source domain, utilizing triplet loss for fine alignment and dynamic time warping-based feature alignment. Additionally, it reduces source-target domain gaps through adversarial training and per-class prototype alignment. Our evaluations on four time-series datasets demonstrate that LogoRA outperforms strong baselines by up to $12.52\%$, showcasing its superiority in time series UDA tasks.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/9e46960ff506775c47d101a6b9c61e32073d772a" target='_blank'>LogoRA: Local-Global Representation Alignment for Robust Time Series Classification</a></td>
+          <td>
+            Huanyu Zhang, Yi-Fan Zhang, Zhang Zhang, Qingsong Wen, Liang Wang
+          </td>
+          <td>2024-09-12</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>4</td>
+        </tr>
+    
+        <tr id="Unlike images and natural language tokens, time series data is highly semantically sparse, resulting in labor-intensive label annotations. Unsupervised and Semi-supervised Domain Adaptation (UDA and SSDA) have demonstrated efficiency in addressing this issue by utilizing pre-labeled source data to train on unlabeled or partially labeled target data. However, in domain adaptation methods designed for downstream classification tasks, directly adapting labeled source samples with unlabelled target samples often results in similar distributions across various classes, thereby compromising the performance of the target classification task. To tackle this challenge, we proposed a Global-Local Alignment Domain Adaptation (GLA-DA) method for multivariate time series data. Data from two domains were initially encoded to align in an intermediate feature space adversarially, achieving Global Feature Alignment (GFA). Subsequently, GLA-DA leveraged the consistency between similarity-based and deep learning-based models to assign pseudo labels to unlabeled target data. This process aims to preserve differences among data with distinct labels by aligning the samples with the same class labels together, achieving Local Class Alignment (LCA). We implemented GLA-DA in both UDA and SSDA scenarios, showcasing its superiority over state-of-the-art methods through extensive experiments on various public datasets. Ablation experiments underscored the significance of key components within GLA-DA.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/89d5e78c21b502378b3cfeadc613d4cfe0046b31" target='_blank'>GLA-DA: Global-Local Alignment Domain Adaptation for Multivariate Time Series</a></td>
+          <td>
+            Gang Tu, Dan Li, Bingxin Lin, Zibin Zheng, See-Kiong Ng
+          </td>
+          <td>2024-10-09</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>1</td>
         </tr>
     
         <tr id="Domain generalization methods aim to learn models robust to domain shift with data from a limited number of source domains and without access to target domain samples during training. Popular domain alignment methods for domain generalization seek to extract domain-invariant features by minimizing the discrepancy between feature distributions across all domains, disregarding inter-domain relationships. In this paper, we instead propose a novel representation learning methodology that selectively enforces prediction consistency between source domains estimated to be closely-related. Specifically, we hypothesize that domains share different class-informative representations, so instead of aligning all domains which can cause negative transfer, we only regularize the discrepancy between closely-related domains. We apply our method to time-series classification tasks and conduct comprehensive experiments on three public real-world datasets. Our method significantly improves over the baseline and achieves better or competitive performance in comparison with state-of-the-art methods in terms of both accuracy and model calibration.">
@@ -98,7 +122,7 @@ hide:
           </td>
           <td>2021-02-13</td>
           <td>ArXiv, DBLP</td>
-          <td>57</td>
+          <td>63</td>
           <td>65</td>
         </tr>
     
diff --git a/docs/recommendations/5d150cec2775f9bc863760448f14104cc8f42368.md b/docs/recommendations/5d150cec2775f9bc863760448f14104cc8f42368.md
index 76f4ec4f..582cde05 100644
--- a/docs/recommendations/5d150cec2775f9bc863760448f14104cc8f42368.md
+++ b/docs/recommendations/5d150cec2775f9bc863760448f14104cc8f42368.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:54 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:41 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2019-03-29</td>
           <td>Proceedings of the National Academy of Sciences of the United States of America</td>
-          <td>615</td>
-          <td>65</td>
+          <td>648</td>
+          <td>67</td>
         </tr>
     
         <tr id="None">
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
     
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>250</td>
-          <td>12</td>
+          <td>289</td>
+          <td>13</td>
         </tr>
     
         <tr id="Discovering the partial differential equations underlying spatio-temporal datasets from very limited and highly noisy observations is of paramount interest in many scientific fields. However, it remains an open question to know when model discovery algorithms based on sparse regression can actually recover the underlying physical processes. In this work, we show the design matrices used to infer the equations by sparse regression can violate the irrepresentability condition (IRC) of the Lasso, even when derived from analytical PDE solutions (i.e. without additional noise). Sparse regression techniques which can recover the true underlying model under violated IRC conditions are therefore required, leading to the introduction of the randomised adaptive Lasso. We show once the latter is integrated within the deep learning model discovery framework DeepMod, a wide variety of nonlinear and chaotic canonical PDEs can be recovered: (1) up to $\mathcal{O}(2)$ higher noise-to-sample ratios than state-of-the-art algorithms, (2) with a single set of hyperparameters, which paves the road towards truly automated model discovery.">
@@ -99,7 +99,7 @@ hide:
           <td>2021-06-22</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
     
         <tr id="Harnessing data to discover the underlying governing laws or equations that describe the behavior of complex physical systems can significantly advance our modeling, simulation and understanding of such systems in various science and engineering disciplines. Recent advances in sparse identification show encouraging success in distilling closed-form governing equations from data for a wide range of nonlinear dynamical systems. However, the fundamental bottleneck of this approach lies in the robustness and scalability with respect to data scarcity and noise. This work introduces a novel physics-informed deep learning framework to discover governing partial differential equations (PDEs) from scarce and noisy data for nonlinear spatiotemporal systems. In particular, this approach seamlessly integrates the strengths of deep neural networks for rich representation learning, automatic differentiation and sparse regression to approximate the solution of system variables, compute essential derivatives, as well as identify the key derivative terms and parameters that form the structure and explicit expression of the PDEs. The efficacy and robustness of this method are demonstrated on discovering a variety of PDE systems with different levels of data scarcity and noise. The resulting computational framework shows the potential for closed-form model discovery in practical applications where large and accurate datasets are intractable to capture.">
@@ -111,7 +111,19 @@ hide:
           <td>2020-05-05</td>
           <td>ArXiv</td>
           <td>19</td>
-          <td>12</td>
+          <td>13</td>
+        </tr>
+    
+        <tr id="The combination of machine learning (ML) and sparsity-promoting techniques is enabling direct extraction of governing equations from data, revolutionizing computational modeling in diverse fields of science and engineering. The discovered dynamical models could be used to address challenges in climate science, neuroscience, ecology, finance, epidemiology, and beyond. However, most existing sparse identification methods for discovering dynamical systems treat the whole system as one without considering the interactions between subsystems. As a result, such models are not able to capture small changes in the emergent system behavior. To address this issue, we developed a new method called Sparse Identification of Nonlinear Dynamical Systems from Graph-structured data (SINDyG), which incorporates the network structure into sparse regression to identify model parameters that explain the underlying network dynamics. SINDyG discovers the governing equations of network dynamics while offering improvements in accuracy and model simplicity.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/4da518ae886702aed27abdd8f94fea3042ab9377" target='_blank'>Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems</a></td>
+          <td>
+            Mohammad Amin Basiri, Sina Khanmohammadi
+          </td>
+          <td>2024-09-02</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>3</td>
         </tr>
     
         <tr id="In this work, we blend machine learning and dictionary-based learning with numerical analysis tools to discover differential equations from noisy and sparsely sampled measurement data of time-dependent processes. We use the fact that given a dictionary containing large candidate nonlinear functions, dynamical models can often be described by a few appropriately chosen basis functions. As a result, we obtain parsimonious models that can be better interpreted by practitioners, and potentially generalize better beyond the sampling regime than black-box modelling. In this work, we integrate a numerical integration framework with dictionary learning that yields differential equations without requiring or approximating derivative information at any stage. Hence, it is utterly effective for corrupted and sparsely sampled data. We discuss its extension to governing equations, containing rational nonlinearities that typically appear in biological networks. Moreover, we generalized the method to governing equations subject to parameter variations and externally controlled inputs. We demonstrate the efficiency of the method to discover a number of diverse differential equations using noisy measurements, including a model describing neural dynamics, chaotic Lorenz model, Michaelis–Menten kinetics and a parameterized Hopf normal form.">
@@ -122,7 +134,7 @@ hide:
           </td>
           <td>2021-05-11</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>38</td>
+          <td>41</td>
           <td>53</td>
         </tr>
     
@@ -135,7 +147,7 @@ hide:
           <td>2014-12-10</td>
           <td>The European Physical Journal Special Topics</td>
           <td>64</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
     
         <tr id="None">
@@ -147,7 +159,7 @@ hide:
           <td>2014-12-01</td>
           <td>The European Physical Journal Special Topics</td>
           <td>7</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c.md b/docs/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c.md
index 4a73e4b2..2c0aec77 100644
--- a/docs/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c.md
+++ b/docs/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:54 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:41 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,20 +50,20 @@ hide:
           </td>
           <td>2022-07-01</td>
           <td>Nature Computational Science</td>
-          <td>69</td>
-          <td>20</td>
+          <td>77</td>
+          <td>21</td>
         </tr>
     
         <tr id="Technological advancements have substantially increased computational power and data availability, enabling the application of powerful machine-learning (ML) techniques across various fields. However, our ability to leverage ML methods for scientific discovery, {\it i.e.} to obtain fundamental and formalized knowledge about natural processes, is still in its infancy. In this review, we explore how the scientific community can increasingly leverage ML techniques to achieve scientific discoveries. We observe that the applicability and opportunity of ML depends strongly on the nature of the problem domain, and whether we have full ({\it e.g.}, turbulence), partial ({\it e.g.}, computational biochemistry), or no ({\it e.g.}, neuroscience) {\it a-priori} knowledge about the governing equations and physical properties of the system. Although challenges remain, principled use of ML is opening up new avenues for fundamental scientific discoveries. Throughout these diverse fields, there is a theme that ML is enabling researchers to embrace complexity in observational data that was previously intractable to classic analysis and numerical investigations.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/f7214e1640b55a726c771c629c873920245e0839" target='_blank'>Opportunities for machine learning in scientific discovery</a></td>
           <td>
-            Ricardo Vinuesa, Jean Rabault, Hossein Azizpour, Stefan Bauer, Bingni W. Brunton, Arne Elofsson, Elias Jarlebring, Hedvig Kjellstrom, Stefano Markidis, David Marlevi, Paola Cinnella, S. Brunton
+            Ricardo Vinuesa, Jean Rabault, H. Azizpour, Stefan Bauer, Bingni W. Brunton, Arne Elofsson, Elias Jarlebring, Hedvig Kjellstrom, Stefano Markidis, David Marlevi, Paola Cinnella, S. Brunton
           </td>
           <td>2024-05-07</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
     
         <tr id="None">
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2019-04-20</td>
           <td>J. Comput. Phys.</td>
-          <td>101</td>
-          <td>34</td>
+          <td>104</td>
+          <td>36</td>
         </tr>
     
         <tr id="None">
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2021-07-22</td>
           <td>Communications Physics</td>
-          <td>17</td>
-          <td>94</td>
+          <td>20</td>
+          <td>95</td>
         </tr>
     
         <tr id="The extraction of models from data (in a sense, the “understanding” of the physical laws giving rise to the data) is a fundamental cognitive as well as scientific challenge. We show a geometric/analytic learning algorithm capable of creating minimal descriptions of parametrically dependent unknown nonlinear dynamical systems. This is accomplished by the data-driven discovery of useful intrinsic-state variables and parameters in terms of which one can empirically model the underlying dynamics. We discuss an informed observation geometry that enables us to formulate models without first principles as well as without closed form equations. (See pp. E7865–E7874.)">
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2020-09-25</td>
           <td>ArXiv</td>
-          <td>88</td>
+          <td>90</td>
           <td>17</td>
         </tr>
     
@@ -134,7 +134,7 @@ hide:
           </td>
           <td>2020-01-30</td>
           <td>J. Comput. Phys.</td>
-          <td>32</td>
+          <td>33</td>
           <td>47</td>
         </tr>
     
@@ -147,8 +147,8 @@ hide:
           </td>
           <td>2020-01-13</td>
           <td>ArXiv</td>
-          <td>493</td>
-          <td>25</td>
+          <td>509</td>
+          <td>26</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37.md b/docs/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37.md
index b0c0c3bf..87487594 100644
--- a/docs/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37.md
+++ b/docs/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:32 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:21 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2023-04-23</td>
           <td>Sensors (Basel, Switzerland)</td>
-          <td>22</td>
+          <td>28</td>
           <td>77</td>
         </tr>
     
@@ -75,7 +75,7 @@ hide:
           <td>2023-06-10</td>
           <td>2023 31st European Signal Processing Conference (EUSIPCO)</td>
           <td>0</td>
-          <td>38</td>
+          <td>39</td>
         </tr>
     
         <tr id="Significant progress has been made in representation learning, especially with recent success on self-supervised contrastive learning. However, for time series with less intuitive or semantic meaning, sampling bias may be inevitably encountered in unsupervised approaches. Although supervised contrastive learning has shown superior performance by leveraging label information, it may also suffer from class collapse. In this study, we consider a realistic scenario in industry with limited annotation information available. A supervised contrastive framework is developed for high-frequency time series representation and classification, wherein a novel variant of supervised contrastive loss is proposed to include multiple augmentations while induce spread within each class. Experiments on four mainstream public datasets as well as a series of sensitivity and ablation analyses demonstrate that the learned representations are effective and robust compared with the direct supervised learning and self-supervised learning, notably under the minimal few-shot situation.">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2023-06-26</td>
           <td>DBLP</td>
-          <td>2</td>
+          <td>4</td>
           <td>140</td>
         </tr>
     
@@ -98,8 +98,8 @@ hide:
           </td>
           <td>2023-10-21</td>
           <td>ArXiv</td>
-          <td>21</td>
-          <td>3</td>
+          <td>30</td>
+          <td>4</td>
         </tr>
     
         <tr id="Unsupervised domain adaptation (UDA) provides a strategy for improving machine learning performance in data-rich (target) domains where ground truth labels are inaccessible but can be found in related (source) domains. In cases where meta-domain information such as label distributions is available, weak supervision can further boost performance. We propose a novel framework, CALDA, to tackle these two problems. CALDA synergistically combines the principles of contrastive learning and adversarial learning to robustly support multi-source UDA (MS-UDA) for time series data. Similar to prior methods, CALDA utilizes adversarial learning to align source and target feature representations. Unlike prior approaches, CALDA additionally leverages cross-source label information across domains. CALDA pulls examples with the same label close to each other, while pushing apart examples with different labels, reshaping the space through contrastive learning. Unlike prior contrastive adaptation methods, CALDA requires neither data augmentation nor pseudo labeling, which may be more challenging for time series. We empirically validate our proposed approach. Based on results from human activity recognition, electromyography, and synthetic datasets, we find utilizing cross-source information improves performance over prior time series and contrastive methods. Weak supervision further improves performance, even in the presence of noise, allowing CALDA to offer generalizable strategies for MS-UDA.">
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2021-09-30</td>
           <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
-          <td>16</td>
+          <td>21</td>
           <td>72</td>
         </tr>
     
@@ -123,7 +123,19 @@ hide:
           <td>2023-08-04</td>
           <td>Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining</td>
           <td>0</td>
-          <td>24</td>
+          <td>25</td>
+        </tr>
+    
+        <tr id="In this work, we investigate the time series representation learning problem using self-supervised techniques. Contrastive learning is well-known in this area as it is a powerful method for extracting information from the series and generating task-appropriate representations. Despite its proficiency in capturing time series characteristics, these techniques often overlook a critical factor - the inherent noise in this type of data, a consideration usually emphasized in general time series analysis. Moreover, there is a notable absence of attention to developing efficient yet lightweight encoder architectures, with an undue focus on delivering contrastive losses. Our work address these gaps by proposing an innovative training strategy that promotes consistent representation learning, accounting for the presence of noise-prone signals in natural time series. Furthermore, we propose an encoder architecture that incorporates dilated convolution within the Inception block, resulting in a scalable and robust network with a wide receptive field. Experimental findings underscore the effectiveness of our method, consistently outperforming state-of-the-art approaches across various tasks, including forecasting, classification, and abnormality detection. Notably, our method attains the top rank in over two-thirds of the classification UCR datasets, utilizing only 40% of the parameters compared to the second-best approach. Our source code for CoInception framework is accessible at https://github.com/anhduy0911/CoInception.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/b1909388ab7a548f6b27dc95a3a279a1a24234fe" target='_blank'>Improving Time Series Encoding with Noise-Aware Self-Supervised Learning and an Efficient Encoder</a></td>
+          <td>
+            Anh Duy Nguyen, Trang H. Tran, Hieu Pham, Phi-Le Nguyen, Lam M. Nguyen
+          </td>
+          <td>2023-06-11</td>
+          <td>ArXiv</td>
+          <td>3</td>
+          <td>20</td>
         </tr>
     
         <tr id="Monitoring and recognizing patterns in continuous sensing data is crucial for many practical applications. These real-world time-series data are often nonstationary, characterized by varying statistical and spectral properties over time. This poses a significant challenge in developing learning models that can effectively generalize across different distributions. In this work, based on our observation that nonstationary statistics are intrinsically linked to the phase information, we propose a time-series learning framework, PhASER. It consists of three novel elements: 1) phase augmentation that diversifies non-stationarity while preserving discriminatory semantics, 2) separate feature encoding by viewing time-varying magnitude and phase as independent modalities, and 3) feature broadcasting by incorporating phase with a novel residual connection for inherent regularization to enhance distribution invariant learning. Upon extensive evaluation on 5 datasets from human activity recognition, sleep-stage classification, and gesture recognition against 10 state-of-the-art baseline methods, we demonstrate that PhASER consistently outperforms the best baselines by an average of 5% and up to 13% in some cases. Moreover, PhASER's principles can be applied broadly to boost the generalization ability of existing time series classification models.">
@@ -135,7 +147,7 @@ hide:
           <td>2024-02-05</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>9</td>
+          <td>10</td>
         </tr>
     
         <tr id="Recent contrastive methods show significant improvement in self-supervised learning in several domains. In particular, contrastive methods are most effective where data augmentation can be easily constructed e.g. in computer vision. However, they are less successful in domains without established data transformations such as time series data. In this paper, we propose a novel self-supervised learning framework that combines contrastive learning with neural processes. It relies on recent advances in neural processes to perform time series forecasting. This allows to generate augmented versions of data by employing a set of various sampling functions and, hence, avoid manually designed augmentations. We extend conventional neural processes and propose a new contrastive loss to learn times series representations in a self-supervised setup. Therefore, unlike previous self-supervised methods, our augmentation pipeline is task-agnostic, enabling our method to perform well across various applications. In particular, a ResNet with a linear classifier trained using our approach is able to outperform state-of-the-art techniques across industrial, medical and audio datasets improving accuracy over 10% in ECG periodic data. We further demonstrate that our self-supervised representations are more efficient in the latent space, improving multiple clustering indexes and that fine-tuning our method on 10% of labels achieves results competitive to fully-supervised learning.">
@@ -150,18 +162,6 @@ hide:
           <td>13</td>
         </tr>
     
-        <tr id="Learning representations from unlabeled time series data is a challenging problem. Most existing self-supervised and unsupervised approaches in the time-series domain fall short in capturing low-and high-frequency features at the same time. As a result, the generalization ability of the learned representations remains limited. Furthermore, some of these methods employ large-scale models like transformers or rely on computationally expensive techniques such as contrastive learning. To tackle these problems, we propose a noncontrastive self-supervised learning (SSL) approach that efficiently captures low-and high-frequency features in a cost-effective manner. The proposed framework comprises a Siamese configuration of a deep neural network with two weight-sharing branches which are followed by low-and high-frequency feature extraction modules. The two branches of the proposed network allow bootstrapping of the latent representation by taking two different augmented views of raw time series data as input. The augmented views are created by applying random transformations sampled from a single set of augmentations. The low-and high-frequency feature extraction modules of the proposed network contain a combination of multilayer perceptron (MLP) and temporal convolutional network (TCN) heads, respectively, which capture the temporal dependencies from the raw input data at various scales due to the varying receptive fields. To demonstrate the robustness of our model, we performed extensive experiments and ablation studies on five real-world time-series datasets. Our method achieves state-of-art performance on all the considered datasets.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4b65c76f7f28e7ba3ea44ebb1b3e4355a148d0de" target='_blank'>Large Scale Time-Series Representation Learning via Simultaneous Low and High Frequency Feature Bootstrapping</a></td>
-          <td>
-            Vandan Gorade, Azad Singh, Deepak Mishra
-          </td>
-          <td>2022-04-24</td>
-          <td>IEEE transactions on neural networks and learning systems</td>
-          <td>5</td>
-          <td>6</td>
-        </tr>
-    
   </tbody>
   <tfoot>
     <tr>
diff --git a/docs/recommendations/73dd9c49f205280991826b2ea4b50344203916b4.md b/docs/recommendations/73dd9c49f205280991826b2ea4b50344203916b4.md
index d60c0826..8dcb7561 100644
--- a/docs/recommendations/73dd9c49f205280991826b2ea4b50344203916b4.md
+++ b/docs/recommendations/73dd9c49f205280991826b2ea4b50344203916b4.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:04 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:49 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2022-09-04</td>
           <td>Scientific Reports</td>
-          <td>2</td>
-          <td>23</td>
+          <td>3</td>
+          <td>24</td>
         </tr>
     
         <tr id="Data-driven modeling frameworks that adopt sparse regression techniques, such as sparse identification of nonlinear dynamics (SINDy) and its modifications, are developed to resolve difficulties in extracting underlying dynamics from experimental data. In contrast to neural-network-based methods, these methods are designed to obtain white-box analytical models. In this work, we incorporate the concept of SINDy and knowledge in the field of classical mechanics to identify interpretable and sparse expressions of total energy and the Lagrangian that shelters the hidden dynamics. Moreover, our method (hereafter referred as Lagrangian-SINDy) is developed to use knowledge of simple systems that form the system being analyzed to ensure the likelihood of correct results and to improve the learning pace. Lagrangian-SINDy is highly accurate in discovering interpretable dynamics via energy-related physical quantities. Its performance is validated with three popular multi-DOF nonlinear dynamical systems, namely the spherical pendulum, double pendulum and cart-pendulum system. Comparisons with other SINDy-based methods are made and Lagrangian-SINDy is found to provide the most compact analytical models.">
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2020-01-31</td>
           <td>IEEE Robotics and Automation Letters</td>
-          <td>23</td>
-          <td>23</td>
+          <td>24</td>
+          <td>24</td>
         </tr>
     
         <tr id="In this work we analyze the effectiveness of the Sparse Identification of Nonlinear Dynamics (SINDy) technique on three benchmark datasets for nonlinear identification, to provide a better understanding of its suitability when tackling real dynamical systems. While SINDy can be an appealing strategy for pursuing physics-based learning, our analysis highlights difficulties in dealing with unobserved states and non-smooth dynamics. Due to the ubiquity of these features in real systems in general, and control applications in particular, we complement our analysis with hands-on approaches to tackle these issues in order to exploit SINDy also in these challenging contexts.">
@@ -75,7 +75,7 @@ hide:
           <td>2024-03-01</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>1</td>
+          <td>2</td>
         </tr>
     
         <tr id="In this thesis, we draw inspiration from both classical system identification and modern machine learning in order to solve estimation problems for real-world, physical systems. The main approach to estimation and learning adopted is optimization based. Concepts such as regularization will be utilized for encoding of prior knowledge and basis-function expansions will be used to add nonlinear modeling power while keeping data requirements practical. The thesis covers a wide range of applications, many inspired by applications within robotics, but also extending outside this already wide field. Usage of the proposed methods and algorithms are in many cases illustrated in the real-world applications that motivated the research. Topics covered include dynamics modeling and estimation, model-based reinforcement learning, spectral estimation, friction modeling and state estimation and calibration in robotic machining. In the work on modeling and identification of dynamics, we develop regularization strategies that allow us to incorporate prior domain knowledge into flexible, overparameterized models. We make use of classical control theory to gain insight into training and regularization while using flexible tools from modern deep learning. A particular focus of the work is to allow use of modern methods in scenarios where gathering data is associated with a high cost. In the robotics-inspired parts of the thesis, we develop methods that are practically motivated and ensure that they are implementable also outside the research setting. We demonstrate this by performing experiments in realistic settings and providing open-source implementations of all proposed methods and algorithms.">
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2021-12-15</td>
           <td>2022 3rd International Conference on Artificial Intelligence, Robotics and Control (AIRC)</td>
-          <td>4</td>
+          <td>5</td>
           <td>5</td>
         </tr>
     
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2017-11-15</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>435</td>
-          <td>65</td>
+          <td>461</td>
+          <td>67</td>
         </tr>
     
         <tr id="None">
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2023-03-01</td>
           <td>Comput. Chem. Eng.</td>
-          <td>12</td>
+          <td>15</td>
           <td>76</td>
         </tr>
     
@@ -134,8 +134,8 @@ hide:
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>202</td>
-          <td>65</td>
+          <td>221</td>
+          <td>67</td>
         </tr>
     
         <tr id="We consider the problem of learning the dynamics of a linear system when one has access to data generated by an auxiliary system that shares similar (but not identical) dynamics, in addition to data from the true system. We use a weighted least squares approach, and provide finite sample error bounds of the learned model as a function of the number of samples and various system parameters from the two systems as well as the weight assigned to the auxiliary data. We show that the auxiliary data can help to reduce the intrinsic system identification error due to noise, at the price of adding a portion of error that is due to the differences between the two system models. We further provide a data-dependent bound that is computable when some prior knowledge about the systems, such as upper bounds on noise levels and model difference, is available. This bound can also be used to determine the weight that should be assigned to the auxiliary data during the model training stage.">
@@ -147,7 +147,7 @@ hide:
           <td>2023-02-08</td>
           <td>ArXiv</td>
           <td>7</td>
-          <td>36</td>
+          <td>37</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/883547fdbd88552328a6615ec620f96e39c57018.md b/docs/recommendations/883547fdbd88552328a6615ec620f96e39c57018.md
index 7ccd73c3..4c26a2c8 100644
--- a/docs/recommendations/883547fdbd88552328a6615ec620f96e39c57018.md
+++ b/docs/recommendations/883547fdbd88552328a6615ec620f96e39c57018.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:07 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:51 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2021-07-05</td>
           <td>Royal Society Open Science</td>
-          <td>54</td>
-          <td>31</td>
+          <td>61</td>
+          <td>32</td>
         </tr>
     
         <tr id="Sparse model identification enables nonlinear dynamical system discovery from data. However, the control of false discoveries for sparse model identification is challenging, especially in the low-data and high-noise limit. In this paper, we perform a theoretical study on ensemble sparse model discovery, which shows empirical success in terms of accuracy and robustness to noise. In particular, we analyse the bootstrapping-based sequential thresholding least-squares estimator. We show that this bootstrapping-based ensembling technique can perform a provably correct variable selection procedure with an exponential convergence rate of the error rate. In addition, we show that the ensemble sparse model discovery method can perform computationally efficient uncertainty estimation, compared to expensive Bayesian uncertainty quantification methods via MCMC. We demonstrate the convergence properties and connection to uncertainty quantification in various numerical studies on synthetic sparse linear regression and sparse model discovery. The experiments on sparse linear regression support that the bootstrapping-based sequential thresholding least-squares method has better performance for sparse variable selection compared to LASSO, thresholding least-squares, and bootstrapping-based LASSO. In the sparse model discovery experiment, we show that the bootstrapping-based sequential thresholding least-squares method can provide valid uncertainty quantification, converging to a delta measure centered around the true value with increased sample sizes. Finally, we highlight the improved robustness to hyperparameter selection under shifting noise and sparsity levels of the bootstrapping-based sequential thresholding least-squares method compared to other sparse regression methods.">
@@ -63,7 +63,7 @@ hide:
           <td>2023-01-30</td>
           <td>ArXiv</td>
           <td>11</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
     
         <tr id="None">
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
     
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>202</td>
-          <td>65</td>
+          <td>221</td>
+          <td>67</td>
         </tr>
     
         <tr id="We propose a fast probabilistic framework for identifying differential equations governing the dynamics of observed data. We recast the SINDy method within a Bayesian framework and use Gaussian approximations for the prior and likelihood to speed up computation. The resulting method, Bayesian-SINDy, not only quantifies uncertainty in the parameters estimated but also is more robust when learning the correct model from limited and noisy data. Using both synthetic and real-life examples such as Lynx-Hare population dynamics, we demonstrate the effectiveness of the new framework in learning correct model equations and compare its computational and data efficiency with existing methods. Because Bayesian-SINDy can quickly assimilate data and is robust against noise, it is particularly suitable for biological data and real-time system identification in control. Its probabilistic framework also enables the calculation of information entropy, laying the foundation for an active learning strategy.">
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
     
         <tr id="The SINDy algorithm has been successfully used to identify the governing equations of dynamical systems from time series data. However, SINDy assumes the user has prior knowledge of the variables in the system and of a function library that can act as a basis for the system. In this paper, we demonstrate on real world data how the Augmented SINDy algorithm outperforms SINDy in the presence of system variable uncertainty. We then show SINDy can be further augmented to perform robustly when both kinds of uncertainty are present.">
@@ -126,6 +126,18 @@ hide:
           <td>0</td>
         </tr>
     
+        <tr id="Active learning algorithms strategically select informative input samples, aiming to enhance understanding and optimize sample efficiency for dynamical systems. The efficacy of machine learning (ML) models with uncertainty quantification in identifying dynamical systems with different characteristics has not yet been explored. This paper investigates four ML models: Gaussian processes (GPs) with radial basis function (RBF), GPs with Matérn kernels, Bayesian neural networks (NNs) with Monte Carlo dropout(MCD), and ensemble neural networks (ENNs). Both continuous and non-continuous learning strategies are employed for all models. Five well known dynamical systems with different characteristics, state dimensionalities, and dynamical complexities are simulated to test the performance of the models along the metrics of test RMSE as well as samples and run time until convergence. Through systematic experimentation, this paper provides guidance for researchers selecting models and algorithms in active learning for system identification, with insights into their strengths and limitations. Gaussian Processes with RBF kernels are shown to be the best overall choice while ENNs with Bayesian inference emerge as a scalable alternative, showing promise for high-dimensional tasks. Employing continuous training has proven to enhance the performance of Gaussian processes with RBF kernels, as well as ENNs, resulting in significant improvements in both sample and run-time efficiency.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/4a3887d48d70464cf825d7d6d8834e36e77b5d04" target='_blank'>Comparative Analysis of Uncertainty Quantification Models in Active Learning for Efficient System Identification of Dynamical Systems</a></td>
+          <td>
+            Hans Mertens, Frances Zhu
+          </td>
+          <td>2024-08-28</td>
+          <td>2024 IEEE 20th International Conference on Automation Science and Engineering (CASE)</td>
+          <td>0</td>
+          <td>0</td>
+        </tr>
+    
         <tr id="We consider the data-driven discovery of governing equations from time-series data in the limit of high noise. The algorithms developed describe an extensive toolkit of methods for circumventing the deleterious effects of noise in the context of the <italic>sparse identification of nonlinear dynamics</italic> (SINDy) framework. We offer two primary contributions, both focused on noisy data acquired from a system <inline-formula> <tex-math notation="LaTeX">$\dot { \boldsymbol x} = { \boldsymbol f} ({ \boldsymbol x})$ </tex-math></inline-formula>. First, we propose, for use in high-noise settings, an extensive toolkit of critically enabling extensions for the SINDy regression method, to progressively cull functionals from an over-complete library and yield a set of sparse equations that regress to the derivate <inline-formula> <tex-math notation="LaTeX">$\dot { \boldsymbol {x}}$ </tex-math></inline-formula>. This toolkit includes: (regression step) weight timepoints based on estimated noise, use ensembles to estimate coefficients, and regress using FFTs; (culling step) leverage linear dependence of functionals, and restore and protect culled functionals based on Figures of Merit (FoMs). In a novel Assessment step, we define FoMs that compare model predictions to the original time-series (i.e., <inline-formula> <tex-math notation="LaTeX">${ \boldsymbol x}(t)$ </tex-math></inline-formula> rather than <inline-formula> <tex-math notation="LaTeX">$\dot { \boldsymbol {x}}(t)$ </tex-math></inline-formula>). These innovations can extract sparse governing equations and coefficients from high-noise time-series data (e.g., 300% added noise). For example, it discovers the correct sparse libraries in the Lorenz system, with median coefficient estimate errors equal to 1%−3% (for 50% noise), 6%−8% (for 100% noise), and 23%−25% (for 300% noise). The enabling modules in the toolkit are combined into a single method, but the individual modules can be tactically applied in other equation discovery methods (SINDy or not) to improve results on high-noise data. Second, we propose a technique, applicable to any model discovery method based on <inline-formula> <tex-math notation="LaTeX">$\dot { \boldsymbol x} = { \boldsymbol f} ({ \boldsymbol x})$ </tex-math></inline-formula>, to assess the accuracy of a discovered model in the context of non-unique solutions due to noisy data. Currently, this non-uniqueness can obscure a discovered model’s accuracy and thus a discovery method’s effectiveness. We describe a technique that uses linear dependencies among functionals to transform a discovered model into an equivalent form that is closest to the <italic>true</italic> model, enabling more accurate assessment of a discovered model’s correctness.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/a9803137946f0dc8997f23de925d5aec6e92dfb9" target='_blank'>A Toolkit for Data-Driven Discovery of Governing Equations in High-Noise Regimes</a></td>
@@ -135,7 +147,7 @@ hide:
           <td>2021-11-08</td>
           <td>IEEE Access</td>
           <td>16</td>
-          <td>31</td>
+          <td>32</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271.md b/docs/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271.md
index a7b002a2..65086187 100644
--- a/docs/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271.md
+++ b/docs/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:57 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:43 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -42,18 +42,6 @@ hide:
   </thead>
   <tbody>
     
-        <tr id="Many dynamical systems of interest are nonlinear, with examples in turbulence, epidemiology, neuroscience, and finance, making them difficult to control using linear approaches. Model predictive control (MPC) is a powerful model-based optimization technique that enables the control of such nonlinear systems with constraints. However, modern systems often lack computationally tractable models, motivating the use of system identification techniques to learn accurate and efficient models for real-time control. In this tutorial article, we review emerging data-driven methods for model discovery and how they are used for nonlinear MPC. In particular, we focus on the sparse identification of nonlinear dynamics (SINDy) algorithm and show how it may be used with MPC on an infectious disease control example. We compare the performance against MPC based on a linear dynamic mode decomposition (DMD) model. Code is provided to run the tutorial examples and may be modified to extend this data-driven control framework to arbitrary nonlinear systems.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/700552dde778bd1f6b76f7237e9b561597b859ba" target='_blank'>SINDy with Control: A Tutorial</a></td>
-          <td>
-            Urban Fasel, E. Kaiser, J. Kutz, Bingni W. Brunton, S. Brunton
-          </td>
-          <td>2021-08-30</td>
-          <td>2021 60th IEEE Conference on Decision and Control (CDC)</td>
-          <td>48</td>
-          <td>65</td>
-        </tr>
-    
         <tr id="In this work we analyze the effectiveness of the Sparse Identification of Nonlinear Dynamics (SINDy) technique on three benchmark datasets for nonlinear identification, to provide a better understanding of its suitability when tackling real dynamical systems. While SINDy can be an appealing strategy for pursuing physics-based learning, our analysis highlights difficulties in dealing with unobserved states and non-smooth dynamics. Due to the ubiquity of these features in real systems in general, and control applications in particular, we complement our analysis with hands-on approaches to tackle these issues in order to exploit SINDy also in these challenging contexts.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/99877062e70fae24c531d58aa880fa443ec10c9b" target='_blank'>SINDy vs Hard Nonlinearities and Hidden Dynamics: a Benchmarking Study</a></td>
@@ -63,7 +51,7 @@ hide:
           <td>2024-03-01</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>1</td>
+          <td>2</td>
         </tr>
     
         <tr id="Data-driven modeling frameworks that adopt sparse regression techniques, such as sparse identification of nonlinear dynamics (SINDy) and its modifications, are developed to resolve difficulties in extracting underlying dynamics from experimental data. In contrast to neural-network-based methods, these methods are designed to obtain white-box analytical models. In this work, we incorporate the concept of SINDy and knowledge in the field of classical mechanics to identify interpretable and sparse expressions of total energy and the Lagrangian that shelters the hidden dynamics. Moreover, our method (hereafter referred as Lagrangian-SINDy) is developed to use knowledge of simple systems that form the system being analyzed to ensure the likelihood of correct results and to improve the learning pace. Lagrangian-SINDy is highly accurate in discovering interpretable dynamics via energy-related physical quantities. Its performance is validated with three popular multi-DOF nonlinear dynamical systems, namely the spherical pendulum, double pendulum and cart-pendulum system. Comparisons with other SINDy-based methods are made and Lagrangian-SINDy is found to provide the most compact analytical models.">
@@ -74,8 +62,8 @@ hide:
           </td>
           <td>2020-01-31</td>
           <td>IEEE Robotics and Automation Letters</td>
-          <td>23</td>
-          <td>23</td>
+          <td>24</td>
+          <td>24</td>
         </tr>
     
         <tr id="Accurately modelling the nonlinear dynamics of a system from measurement data is a challenging yet vital topic. The sparse identification of nonlinear dynamics (SINDy) algorithm is one approach to discover dynamical systems models from data. Although extensions have been developed to identify implicit dynamics, or dynamics described by rational functions, these extensions are extremely sensitive to noise. In this work, we develop SINDy-PI (parallel, implicit), a robust variant of the SINDy algorithm to identify implicit dynamics and rational nonlinearities. The SINDy-PI framework includes multiple optimization algorithms and a principled approach to model selection. We demonstrate the ability of this algorithm to learn implicit ordinary and partial differential equations and conservation laws from limited and noisy data. In particular, we show that the proposed approach is several orders of magnitude more noise robust than previous approaches, and may be used to identify a class of ODE and PDE dynamics that were previously unattainable with SINDy, including for the double pendulum dynamics and simplified model for the Belousov–Zhabotinsky (BZ) reaction.">
@@ -86,8 +74,8 @@ hide:
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>202</td>
-          <td>65</td>
+          <td>221</td>
+          <td>67</td>
         </tr>
     
         <tr id="Modern societies have an abundance of data yet good system models are rare. Unfortunately, many of the current system identification and machine learning techniques fail to generalize outside of the training set, producing models that violate basic physical laws. This work proposes a novel method for the Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI). SINDy-SI is an iterative method that uses Sum-of-Squares (SOS) programming to learn optimally fitted models while guaranteeing that the learned model satisfies side information, such as symmetries and physical laws. Guided by the principle of Occam's razor, that the simplest or most regularized best fitted model is typically the superior choice, during each iteration SINDy-SI prunes the basis functions associated with small coefficients, yielding a sparse dynamical model upon termination. Through several numerical experiments we will show how the combination of side information constraints and sparse polynomial representation cultivates dynamical models that obey known physical laws while displaying impressive generalized performance beyond the training set.">
@@ -102,64 +90,64 @@ hide:
           <td>1</td>
         </tr>
     
-        <tr id="Automated data-driven modeling, the process of directly discovering the governing equations of a system from data, is increasingly being used across the scientific community. PySINDy is a Python package that provides tools for applying the sparse identification of nonlinear dynamics (SINDy) approach to data-driven model discovery. In this major update to PySINDy, we implement several advanced features that enable the discovery of more general differential equations from noisy and limited data. The library of candidate terms is extended for the identification of actuated systems, partial differential equations (PDEs), and implicit differential equations. Robust formulations, including the integral form of SINDy and ensembling techniques, are also implemented to improve performance for real-world data. Finally, we provide a range of new optimization algorithms, including several sparse regression techniques and algorithms to enforce and promote inequality constraints and stability. Together, these updates enable entirely new SINDy model discovery capabilities that have not been reported in the literature, such as constrained PDE identification and ensembling with different sparse regression optimizers.">
+        <tr id="The sparse identification of nonlinear dynamics (SINDy) has been established as an effective technique to produce interpretable models of dynamical systems from time-resolved state data via sparse regression. However, to model parameterized systems, SINDy requires data from transient trajectories for various parameter values over the range of interest, which are typically difficult to acquire experimentally. In this work, we extend SINDy to be able to leverage data on fixed points and/or limit cycles to reduce the number of transient trajectories needed for successful system identification. To achieve this, we incorporate the data on these attractors at various parameter values as constraints in the optimization problem. First, we show that enforcing these as hard constraints leads to an ill-conditioned regression problem due to the large number of constraints. Instead, we implement soft constraints by modifying the cost function to be minimized. This leads to the formulation of a multi-objective sparse regression problem where we simultaneously seek to minimize the error of the fit to the transients trajectories and to the data on attractors, while penalizing the number of terms in the model. Our extension, demonstrated on several numerical examples, is more robust to noisy measurements and requires substantially less training data than the original SINDy method to correctly identify a parameterized dynamical system.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/433ba35ced5d3f47fa2688c81a60363a74193449" target='_blank'>PySINDy: A comprehensive Python package for robust sparse system identification</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/0c03c126b4d641a81099470f03a7d5215a2a6820" target='_blank'>Multi-objective SINDy for parameterized model discovery from single transient trajectory data</a></td>
           <td>
-            A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton
+            Javier A. Lemus, Benjamin Herrmann
           </td>
-          <td>2021-11-12</td>
-          <td>J. Open Source Softw.</td>
-          <td>111</td>
-          <td>65</td>
+          <td>2024-05-14</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>0</td>
         </tr>
     
-        <tr id="None">
+        <tr id="Significance Understanding dynamic constraints and balances in nature has facilitated rapid development of knowledge and enabled technology, including aircraft, combustion engines, satellites, and electrical power. This work develops a novel framework to discover governing equations underlying a dynamical system simply from data measurements, leveraging advances in sparsity techniques and machine learning. The resulting models are parsimonious, balancing model complexity with descriptive ability while avoiding overfitting. There are many critical data-driven problems, such as understanding cognition from neural recordings, inferring climate patterns, determining stability of financial markets, predicting and suppressing the spread of disease, and controlling turbulence for greener transportation and energy. With abundant data and elusive laws, data-driven discovery of dynamics will continue to play an important role in these efforts. Extracting governing equations from data is a central challenge in many diverse areas of science and engineering. Data are abundant whereas models often remain elusive, as in climate science, neuroscience, ecology, finance, and epidemiology, to name only a few examples. In this work, we combine sparsity-promoting techniques and machine learning with nonlinear dynamical systems to discover governing equations from noisy measurement data. The only assumption about the structure of the model is that there are only a few important terms that govern the dynamics, so that the equations are sparse in the space of possible functions; this assumption holds for many physical systems in an appropriate basis. In particular, we use sparse regression to determine the fewest terms in the dynamic governing equations required to accurately represent the data. This results in parsimonious models that balance accuracy with model complexity to avoid overfitting. We demonstrate the algorithm on a wide range of problems, from simple canonical systems, including linear and nonlinear oscillators and the chaotic Lorenz system, to the fluid vortex shedding behind an obstacle. The fluid example illustrates the ability of this method to discover the underlying dynamics of a system that took experts in the community nearly 30 years to resolve. We also show that this method generalizes to parameterized systems and systems that are time-varying or have external forcing.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/899995e30820d657b2e2feb791025cd6b2027e22" target='_blank'>Discovering sparse interpretable dynamics from partial observations</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/5d150cec2775f9bc863760448f14104cc8f42368" target='_blank'>Discovering governing equations from data by sparse identification of nonlinear dynamical systems</a></td>
           <td>
-            Peter Y. Lu, Joan Ariño Bernad, M. Soljačić
+            S. Brunton, J. Proctor, J. Kutz
           </td>
-          <td>2021-07-22</td>
-          <td>Communications Physics</td>
-          <td>17</td>
-          <td>94</td>
+          <td>2015-09-11</td>
+          <td>Proceedings of the National Academy of Sciences</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
     
-        <tr id="Authors of papers retaincopyright and release the workunder a Creative CommonsAttribution 4.0 InternationalLicense (CC-BY)">
+        <tr id="Automated data-driven modeling, the process of directly discovering the governing equations of a system from data, is increasingly being used across the scientific community. PySINDy is a Python package that provides tools for applying the sparse identification of nonlinear dynamics (SINDy) approach to data-driven model discovery. In this major update to PySINDy, we implement several advanced features that enable the discovery of more general differential equations from noisy and limited data. The library of candidate terms is extended for the identification of actuated systems, partial differential equations (PDEs), and implicit differential equations. Robust formulations, including the integral form of SINDy and ensembling techniques, are also implemented to improve performance for real-world data. Finally, we provide a range of new optimization algorithms, including several sparse regression techniques and algorithms to enforce and promote inequality constraints and stability. Together, these updates enable entirely new SINDy model discovery capabilities that have not been reported in the literature, such as constrained PDE identification and ensembling with different sparse regression optimizers.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/85d5e9aea7235bf6aa67d5febc4013923a77d4de" target='_blank'>PySINDy: A Python package for the sparse identification of nonlinear dynamical systems from data</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/433ba35ced5d3f47fa2688c81a60363a74193449" target='_blank'>PySINDy: A comprehensive Python package for robust sparse system identification</a></td>
           <td>
-            Brian M. de Silva, Kathleen P. Champion, M. Quade, Jean-Christophe Loiseau, J. Kutz, S. Brunton
+            A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton
           </td>
-          <td>2020-05-18</td>
+          <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>129</td>
-          <td>65</td>
+          <td>119</td>
+          <td>67</td>
         </tr>
     
-        <tr id="The vast majority of systems of practical interest are characterised by nonlinear dynamics. This renders the control and optimization of such systems a complex task due to their nonlinear behaviour. Additionally, standard methods such as linearizing around a fixed point may not be an effective strategy for many systems, thus requiring an alternative approach. For this reason, we propose a new deep learning framework to discover a transformation of a nonlinear dynamical system to an equivalent higher dimensional linear representation. We demonstrate that the resulting learned linear representation accurately captures the dynamics of the original system for a wider range of conditions than standard linearization. As a result of this, we show that the learned linear model can subsequently be used for the successful control of the original system. We demonstrate this by applying the proposed framework to two examples; one from the literature and a more complex example in the form of a Continuous Stirred Tank Reactor (CSTR).">
+        <tr id="Identifying dynamical systems characterized by nonlinear parameters presents significant challenges in deriving mathematical models that enhance understanding of physics. Traditional methods, such as Sparse Identification of Nonlinear Dynamics (SINDy) and symbolic regression, can extract governing equations from observational data; however, they also come with distinct advantages and disadvantages. This paper introduces a novel method within the SINDy framework, termed ADAM-SINDy, which synthesizes the strengths of established approaches by employing the ADAM optimization algorithm. This facilitates the simultaneous optimization of nonlinear parameters and coefficients associated with nonlinear candidate functions, enabling precise parameter estimation without requiring prior knowledge of nonlinear characteristics such as trigonometric frequencies, exponential bandwidths, or polynomial exponents, thereby addressing a key limitation of SINDy. Through an integrated global optimization, ADAM-SINDy dynamically adjusts all unknown variables in response to data, resulting in an adaptive identification procedure that reduces the sensitivity to the library of candidate functions. The performance of the ADAM-SINDy methodology is demonstrated across a spectrum of dynamical systems, including benchmark coupled nonlinear ordinary differential equations such as oscillators, chaotic fluid flows, reaction kinetics, pharmacokinetics, as well as nonlinear partial differential equations (wildfire transport). The results demonstrate significant improvements in identifying parameterized dynamical systems and underscore the importance of concurrently optimizing all parameters, particularly those characterized by nonlinear parameters. These findings highlight the potential of ADAM-SINDy to extend the applicability of the SINDy framework in addressing more complex challenges in dynamical system identification.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c309d30d468a48f5c5d40e402af5881451bb6117" target='_blank'>Learning Linear Representations of Nonlinear Dynamics Using Deep Learning</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/0c327ed253a2c7b658940860bcf0a9e077d634b6" target='_blank'>ADAM-SINDy: An Efficient Optimization Framework for Parameterized Nonlinear Dynamical System Identification</a></td>
           <td>
-            Akhil Ahmed, E. A. Rio-Chanona, Mehmet Mercangöz
+            Siva Viknesh, Younes Tatari, Amirhossein Arzani
           </td>
-          <td>2022-04-03</td>
+          <td>2024-10-21</td>
           <td>ArXiv</td>
-          <td>3</td>
-          <td>19</td>
+          <td>0</td>
+          <td>24</td>
         </tr>
     
-        <tr id="None">
+        <tr id="We develop a principled mathematical framework for controlling nonlinear, networked dynamical systems. Our method integrates dimensionality reduction, bifurcation theory, and emerging model discovery tools to find low-dimensional subspaces where feed-forward control can be used to manipulate a system to a desired outcome. The method leverages the fact that many high-dimensional networked systems have many fixed points, allowing for the computation of control signals that will move the system between any pair of fixed points. The sparse identification of nonlinear dynamics (SINDy) algorithm is used to fit a nonlinear dynamical system to the evolution on the dominant, low-rank subspace. This then allows us to use bifurcation theory to find collections of constant control signals that will produce the desired objective path for a prescribed outcome. Specifically, we can destabilize a given fixed point while making the target fixed point an attractor. The discovered control signals can be easily projected back to the original high-dimensional state and control space. We illustrate our nonlinear control procedure on established bistable, low-dimensional biological systems, showing how control signals are found that generate switches between the fixed points. We then demonstrate our control procedure for high-dimensional systems on random high-dimensional networks and Hopfield memory networks.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/acd85b948087067938680356a0cb7597de12a25d" target='_blank'>Data-based modeling and control of nonlinear process systems using sparse identification: An overview of recent results</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/4b849d45b437a09d1ea5321a304c5bc411bf3de4" target='_blank'>Nonlinear Control of Networked Dynamical Systems</a></td>
           <td>
-            Fahim Abdullah, P. Christofides
+            Megan Morrison, Nathan Kutz
           </td>
-          <td>2023-03-01</td>
-          <td>Comput. Chem. Eng.</td>
-          <td>12</td>
-          <td>76</td>
+          <td>2020-06-09</td>
+          <td>IEEE Transactions on Network Science and Engineering</td>
+          <td>10</td>
+          <td>5</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da.md b/docs/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da.md
index 7a670f12..052c808a 100644
--- a/docs/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da.md
+++ b/docs/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:08 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:54 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -51,7 +51,7 @@ hide:
           <td>2018-07-14</td>
           <td>ArXiv</td>
           <td>9</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
     
         <tr id="Sparse regression models are increasingly prevalent due to their ease of interpretability and superior out-of-sample performance. However, the exact model of sparse regression with an $\ell_0$ constraint restricting the support of the estimators is a challenging (\NP-hard) non-convex optimization problem. In this paper, we derive new strong convex relaxations for sparse regression. These relaxations are based on the ideal (convex-hull) formulations for rank-one quadratic terms with indicator variables. The new relaxations can be formulated as semidefinite optimization problems in an extended space and are stronger and more general than the state-of-the-art formulations, including the perspective reformulation and formulations with the reverse Huber penalty and the minimax concave penalty functions. Furthermore, the proposed rank-one strengthening can be interpreted as a \textit{non-separable, non-convex, unbiased} sparsity-inducing regularizer, which dynamically adjusts its penalty according to the shape of the error function without inducing bias for the sparse solutions. In our computational experiments with benchmark datasets, the proposed conic formulations are solved within seconds and result in near-optimal solutions (with 0.4\% optimality gap) for non-convex $\ell_0$-problems. Moreover, the resulting estimators also outperform alternative convex approaches from a statistical perspective, achieving high prediction accuracy and good interpretability.">
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2019-01-29</td>
           <td>ArXiv</td>
-          <td>49</td>
-          <td>35</td>
+          <td>48</td>
+          <td>36</td>
         </tr>
     
         <tr id="In the context of sparse recovery, it is known that most of existing regularizers such as $\ell_1$ suffer from some bias incurred by some leading entries (in magnitude) of the associated vector. To neutralize this bias, we propose a class of models with partial regularizers for recovering a sparse solution of a linear system. We show that every local minimizer of these models is sufficiently sparse or the magnitude of all its nonzero entries is above a uniform constant depending only on the data of the linear system. Moreover, for a class of partial regularizers, any global minimizer of these models is a sparsest solution to the linear system. We also establish some sufficient conditions for local or global recovery of the sparsest solution to the linear system, among which one of the conditions is weaker than the best known restricted isometry property (RIP) condition for sparse recovery by $\ell_1$. In addition, a first-order feasible augmented Lagrangian (FAL) method is proposed for solving these models, in which each subproblem is solved by a nonmonotone proximal gradient (NPG) method. Despite the complication of the partial regularizers, we show that each proximal subproblem in NPG can be solved as a certain number of one-dimensional optimization problems, which usually have a closed-form solution. We also show that any accumulation point of the sequence generated by FAL is a first-order stationary point of the models. Numerical results on compressed sensing and sparse logistic regression demonstrate that the proposed models substantially outperform the widely used ones in the literature in terms of solution quality.">
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2015-11-23</td>
           <td>ArXiv</td>
-          <td>36</td>
+          <td>37</td>
           <td>32</td>
         </tr>
     
@@ -113,7 +113,7 @@ As a simple consequence, our approach also provides a new way for estimating $\t
           <td>2021-10-24</td>
           <td>ArXiv</td>
           <td>2</td>
-          <td>16</td>
+          <td>17</td>
         </tr>
     
         <tr id="This paper addresses the robust reconstruction problem of a sparse signal from compressed measurements. We propose a robust formulation for sparse reconstruction that employs the <inline-formula><tex-math notation="LaTeX">$\ell _1$ </tex-math></inline-formula>-norm as the loss function for the residual error and utilizes a generalized nonconvex penalty for sparsity inducing. The <inline-formula><tex-math notation="LaTeX">$\ell _1$</tex-math></inline-formula> -loss is less sensitive to outliers in the measurements than the popular <inline-formula><tex-math notation="LaTeX"> $\ell _2$</tex-math></inline-formula>-loss, while the nonconvex penalty has the capability of ameliorating the bias problem of the popular convex LASSO penalty and thus can yield more accurate recovery. To solve this nonconvex and nonsmooth minimization formulation efficiently, we propose a first-order algorithm based on alternating direction method of multipliers. A smoothing strategy on the <inline-formula><tex-math notation="LaTeX">$\ell _1$</tex-math> </inline-formula>-loss function has been used in deriving the new algorithm to make it convergent. Further, a sufficient condition for the convergence of the new algorithm has been provided for generalized nonconvex regularization. In comparison with several state-of-the-art algorithms, the new algorithm showed better performance in numerical experiments in recovering sparse signals and compressible images. The new algorithm scales well for large-scale problems, as often encountered in image processing.">
@@ -124,8 +124,8 @@ As a simple consequence, our approach also provides a new way for estimating $\t
           </td>
           <td>2017-03-23</td>
           <td>IEEE Transactions on Computational Imaging</td>
-          <td>85</td>
-          <td>30</td>
+          <td>87</td>
+          <td>31</td>
         </tr>
     
         <tr id="None">
diff --git a/docs/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524.md b/docs/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524.md
index 67f55ae5..c3afc5e2 100644
--- a/docs/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524.md
+++ b/docs/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:33 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:25 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2022-04-23</td>
           <td>ArXiv</td>
-          <td>5</td>
+          <td>7</td>
           <td>50</td>
         </tr>
     
@@ -61,8 +61,8 @@ hide:
             Ming Jin, Huan Yee Koh, Qingsong Wen, Daniele Zambon, C. Alippi, G. I. Webb, Irwin King, Shirui Pan
           </td>
           <td>2023-07-07</td>
-          <td>IEEE transactions on pattern analysis and machine intelligence</td>
-          <td>64</td>
+          <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
+          <td>82</td>
           <td>50</td>
         </tr>
     
@@ -75,7 +75,7 @@ hide:
           <td>2023-12-27</td>
           <td>2024 IEEE 40th International Conference on Data Engineering (ICDE)</td>
           <td>0</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
     
         <tr id="None">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2023-09-25</td>
           <td>International Journal of Data Science and Analytics</td>
-          <td>0</td>
+          <td>2</td>
           <td>6</td>
         </tr>
     
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2023-02-22</td>
           <td>ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)</td>
-          <td>5</td>
+          <td>7</td>
           <td>42</td>
         </tr>
     
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2022-06-30</td>
           <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
-          <td>7</td>
-          <td>21</td>
+          <td>8</td>
+          <td>22</td>
         </tr>
     
         <tr id="Multivariate time series forecasting has long received significant attention in real-world applications, such as energy consumption and traffic prediction. While recent methods demonstrate good forecasting abilities, they have three fundamental limitations. (i). <italic>Discrete neural architectures:</italic> Interlacing individually parameterized spatial and temporal blocks to encode rich underlying patterns leads to discontinuous latent state trajectories and higher forecasting numerical errors. (ii). <italic>High complexity:</italic> Discrete approaches complicate models with dedicated designs and redundant parameters, leading to higher computational and memory overheads. (iii). <italic>Reliance on graph priors:</italic> Relying on predefined static graph structures limits their effectiveness and practicability in real-world applications. In this paper, we address all the above limitations by proposing a continuous model to forecast <underline><bold>M</bold></underline>ultivariate <underline><bold>T</bold></underline>ime series with dynamic <underline><bold>G</bold></underline>raph neural <underline><bold>O</bold></underline>rdinary <underline><bold>D</bold></underline>ifferential <underline><bold>E</bold></underline>quations (<monospace>MTGODE</monospace>). Specifically, we first abstract multivariate time series into dynamic graphs with time-evolving node features and unknown graph structures. Then, we design and solve a neural ODE to complement missing graph topologies and unify both spatial and temporal message passing, allowing deeper graph propagation and fine-grained temporal information aggregation to characterize stable and precise latent spatial-temporal dynamics. Our experiments demonstrate the superiorities of <monospace>MTGODE</monospace> from various perspectives on five time series benchmark datasets.">
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2022-02-17</td>
           <td>IEEE Transactions on Knowledge and Data Engineering</td>
-          <td>60</td>
-          <td>43</td>
+          <td>69</td>
+          <td>44</td>
         </tr>
     
         <tr id="Existing multivariate time series anomaly detection methods utilize graph neural networks to model inter-sensor dependencies and achieve satisfactory results. However, these methods often employ a predefined or self-learning static adjacency matrix for graph information aggregation, disregarding the dynamic characteristics of inter-sensor dependencies in the real world, which change over time and are observable at different time scales as dynamic changing graph structure information. Furthermore, existing techniques struggle to capture long-term temporal contextual relationships. To address these issues, we propose a Dynamic Graph Learning with Long and Short-term time model (DGL-LS) for multivariate time series anomaly detection. The network integrates a dynamic graph structure learning method among sensors in multi-time mode and long- and short-term temporal context information to enhance the prediction of multivariate time series anomalies. We propose a strategy for learning dynamic graphs that aggregates sensor node representations at various time intervals, and then uses an attention-based gating unit to evolve the node dynamic characterization and construct a series of adjacency matrices that describe the dynamic changes in graph structure over time. Moreover, we formulate a long-short time convolution to capture the long-term and short-term temporal dependencies within the time series. We demonstrate the superiority of our approach over other state-of-the-art methods by conducting extensive experiments on three publicly available datasets.">
diff --git a/docs/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a.md b/docs/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a.md
index f378fe9a..54452b13 100644
--- a/docs/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a.md
+++ b/docs/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:16 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:11 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>54</td>
+          <td>1</td>
+          <td>55</td>
         </tr>
     
         <tr id="Accurate forecasting of multivariate time series is an extensively studied subject in finance, transportation, and computer science. Fully mining the correlation and causation between the variables in a multivariate time series exhibits noticeable results in improving the performance of a time series model. Recently, some models have explored the dependencies between variables through end-to-end graph structure learning without the need for predefined graphs. However, current models do not incorporate the trade-off between efficiency and flexibility and lack the guidance of domain knowledge in the design of graph structure learning algorithms. This paper alleviates the above issues by proposing Balanced Graph Structure Learning for Forecasting (BGSLF), a novel deep learning model that joins graph structure learning and forecasting. Technically, BGSLF leverages the spatial information into convolutional operations and extracts temporal dynamics using the diffusion convolutional recurrent network. The proposed framework balance the trade-off between efficiency and flexibility by introducing Multi-Graph Generation Network (MGN) and Graph Selection Module. In addition, a method named Smooth Sparse Unit (SSU) is designed to sparse the learned graph structures, which conforms to the sparse spatial correlations in the real world. Extensive experiments on four real-world datasets demonstrate that our model achieves state-of-the-art performances with minor trainable parameters. Code will be made publicly available.">
@@ -63,7 +63,7 @@ hide:
           <td>2022-01-24</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>10</td>
+          <td>11</td>
         </tr>
     
         <tr id="Outstanding achievements of graph neural networks for spatiotemporal time series analysis show that relational constraints introduce an effective inductive bias into neural forecasting architectures. Often, however, the relational information characterizing the underlying data-generating process is unavailable and the practitioner is left with the problem of inferring from data which relational graph to use in the subsequent processing stages. We propose novel, principled - yet practical - probabilistic score-based methods that learn the relational dependencies as distributions over graphs while maximizing end-to-end the performance at task. The proposed graph learning framework is based on consolidated variance reduction techniques for Monte Carlo score-based gradient estimation, is theoretically grounded, and, as we show, effective in practice. In this paper, we focus on the time series forecasting problem and show that, by tailoring the gradient estimators to the graph learning problem, we are able to achieve state-of-the-art performance while controlling the sparsity of the learned graph and the computational scalability. We empirically assess the effectiveness of the proposed method on synthetic and real-world benchmarks, showing that the proposed solution can be used as a stand-alone graph identification procedure as well as a graph learning component of an end-to-end forecasting architecture.">
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2022-05-26</td>
           <td>J. Mach. Learn. Res.</td>
-          <td>11</td>
+          <td>14</td>
           <td>50</td>
         </tr>
     
@@ -98,8 +98,8 @@ hide:
           </td>
           <td>2020-05-24</td>
           <td>Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining</td>
-          <td>981</td>
-          <td>55</td>
+          <td>1082</td>
+          <td>57</td>
         </tr>
     
         <tr id="Graph Neural Networks (GNN) have gained significant traction in the forecasting domain, especially for their capacity to simultaneously account for intra-series temporal correlations and inter-series relationships. This paper introduces a novel Hierarchical GNN (DeepHGNN) framework, explicitly designed for forecasting in complex hierarchical structures. The uniqueness of DeepHGNN lies in its innovative graph-based hierarchical interpolation and an end-to-end reconciliation mechanism. This approach ensures forecast accuracy and coherence across various hierarchical levels while sharing signals across them, addressing a key challenge in hierarchical forecasting. A critical insight in hierarchical time series is the variance in forecastability across levels, with upper levels typically presenting more predictable components. DeepHGNN capitalizes on this insight by pooling and leveraging knowledge from all hierarchy levels, thereby enhancing the overall forecast accuracy. Our comprehensive evaluation set against several state-of-the-art models confirm the superior performance of DeepHGNN. This research not only demonstrates DeepHGNN's effectiveness in achieving significantly improved forecast accuracy but also contributes to the understanding of graph-based methods in hierarchical time series forecasting.">
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2023-11-10</td>
           <td>ArXiv</td>
-          <td>32</td>
+          <td>49</td>
           <td>6</td>
         </tr>
     
diff --git a/docs/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9.md b/docs/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9.md
index c154d345..de19de59 100644
--- a/docs/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9.md
+++ b/docs/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:08 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:07 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,10 +50,22 @@ hide:
           </td>
           <td>2023-12-23</td>
           <td>ACM Transactions on Knowledge Discovery from Data</td>
-          <td>1</td>
+          <td>3</td>
           <td>15</td>
         </tr>
     
+        <tr id="In recent years, spatio-temporal graph neural networks (GNNs) have attracted considerable interest in the field of time series analysis, due to their ability to capture dependencies among variables and across time points. The objective of the presented systematic literature review is hence to provide a comprehensive overview of the various modeling approaches and application domains of GNNs for time series classification and forecasting. A database search was conducted, and over 150 journal papers were selected for a detailed examination of the current state-of-the-art in the field. This examination is intended to offer to the reader a comprehensive collection of proposed models, links to related source code, available datasets, benchmark models, and fitting results. All this information is hoped to assist researchers in future studies. To the best of our knowledge, this is the first systematic literature review presenting a detailed comparison of the results of current spatio-temporal GNN models in different domains. In addition, in its final part this review discusses current limitations and challenges in the application of spatio-temporal GNNs, such as comparability, reproducibility, explainability, poor information capacity, and scalability.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/e6544d542d13247f620834689ca2378cb7fb04f8" target='_blank'>A Systematic Literature Review of Spatio-Temporal Graph Neural Network Models for Time Series Forecasting and Classification</a></td>
+          <td>
+            Flavio Corradini, Marco Gori, Carlo Lucheroni, Marco Piangerelli, Martina Zannotti
+          </td>
+          <td>2024-10-29</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>12</td>
+        </tr>
+    
         <tr id="Time series forecasting lies at the core of important real-world applications in many fields of science and engineering. The abundance of large time series datasets that consist of complex patterns and long-term dependencies has led to the development of various neural network architectures. Graph neural network approaches, which jointly learn a graph structure based on the correlation of raw values of multivariate time series while forecasting, have recently seen great success. However, such solutions are often costly to train and difficult to scale. In this paper, we propose TimeGNN, a method that learns dynamic temporal graph representations that can capture the evolution of inter-series patterns along with the correlations of multiple series. TimeGNN achieves inference times 4 to 80 times faster than other state-of-the-art graph-based methods while achieving comparable forecasting performance">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/d9916caa8da69496ecc8e0b1c229c73a60d6f861" target='_blank'>TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting</a></td>
@@ -62,8 +74,8 @@ hide:
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>54</td>
+          <td>1</td>
+          <td>55</td>
         </tr>
     
         <tr id="Graph-based deep learning methods have become popular tools to process collections of correlated time series. Differently from traditional multivariate forecasting methods, neural graph-based predictors take advantage of pairwise relationships by conditioning forecasts on a (possibly dynamic) graph spanning the time series collection. The conditioning can take the form of an architectural inductive bias on the neural forecasting architecture, resulting in a family of deep learning models called spatiotemporal graph neural networks. Such relational inductive biases enable the training of global forecasting models on large time-series collections, while at the same time localizing predictions w.r.t. each element in the set (i.e., graph nodes) by accounting for local correlations among them (i.e., graph edges). Indeed, recent theoretical and practical advances in graph neural networks and deep learning for time series forecasting make the adoption of such processing frameworks appealing and timely. However, most of the studies in the literature focus on proposing variations of existing neural architectures by taking advantage of modern deep learning practices, while foundational and methodological aspects have not been subject to systematic investigation. To fill the gap, this paper aims to introduce a comprehensive methodological framework that formalizes the forecasting problem and provides design principles for graph-based predictive models and methods to assess their performance. At the same time, together with an overview of the field, we provide design guidelines, recommendations, and best practices, as well as an in-depth discussion of open challenges and future research directions.">
@@ -74,7 +86,7 @@ hide:
           </td>
           <td>2023-10-24</td>
           <td>ArXiv</td>
-          <td>6</td>
+          <td>9</td>
           <td>50</td>
         </tr>
     
@@ -86,7 +98,7 @@ hide:
           </td>
           <td>2023-01-31</td>
           <td>ArXiv</td>
-          <td>3</td>
+          <td>2</td>
           <td>15</td>
         </tr>
     
@@ -110,7 +122,7 @@ hide:
           </td>
           <td>2023-11-10</td>
           <td>ArXiv</td>
-          <td>32</td>
+          <td>49</td>
           <td>6</td>
         </tr>
     
@@ -134,7 +146,7 @@ hide:
           </td>
           <td>2023-02-04</td>
           <td>ArXiv</td>
-          <td>4</td>
+          <td>5</td>
           <td>22</td>
         </tr>
     
@@ -146,20 +158,8 @@ hide:
           </td>
           <td>2024-01-25</td>
           <td>Inf. Sci.</td>
-          <td>0</td>
           <td>1</td>
-        </tr>
-    
-        <tr id="Spatio-temporal forecasting of future values of spatially correlated time series is important across many cyber-physical systems (CPS). Recent studies offer evidence that the use of graph neural networks to capture latent correlations between time series holds a potential for enhanced forecasting. However, most existing methods rely on predefined or self-learning graphs, which are either static or unintentionally dynamic, and thus cannot model the time-varying correlations that exhibit trends and periodicities caused by the regularity of the underlying processes in CPS. To tackle such limitation, we propose Time-aware Graph Structure Learning (TagSL), which extracts time-aware correlations among time series by measuring the interaction of node and time representations in high-dimensional spaces. Notably, we introduce time discrepancy learning that utilizes contrastive learning with distance-based regularization terms to constrain learned spatial correlations to a trend sequence. Additionally, we propose a periodic discriminant function to enable the capture of periodic changes from the state of nodes. Next, we present a Graph Convolution-based Gated Recurrent Unit (GCGRU) that jointly captures spatial and temporal dependencies while learning time-aware and node-specific patterns. Finally, we introduce a unified framework named Time-aware Graph Convolutional Recurrent Network (TGCRN), combining TagSL, and GCGRU in an encoder-decoder architecture for multi-step spatiotemporal forecasting. We report on experiments with TGCRN and popular existing approaches on five real-world datasets, thus providing evidence that TGCRN is capable of advancing the state-of-the-art. We also cover a detailed ablation study and visualization analysis, offering detailed insight into the effectiveness of time-aware structure learning.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2e635d42e768aec7be29fbd203a605c1f011a44d" target='_blank'>Learning Time-Aware Graph Structures for Spatially Correlated Time Series Forecasting</a></td>
-          <td>
-            Minbo Ma, Jilin Hu, Christian S. Jensen, Fei Teng, Peng Han, Zhiqiang Xu, Tian-Jie Li
-          </td>
-          <td>2023-12-27</td>
-          <td>2024 IEEE 40th International Conference on Data Engineering (ICDE)</td>
-          <td>0</td>
-          <td>12</td>
+          <td>1</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419.md b/docs/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419.md
index 39f4a60b..3833a218 100644
--- a/docs/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419.md
+++ b/docs/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:15 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:10 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2023-10-24</td>
           <td>ArXiv</td>
-          <td>6</td>
+          <td>9</td>
           <td>50</td>
         </tr>
     
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2022-09-14</td>
           <td>ArXiv</td>
-          <td>30</td>
+          <td>34</td>
           <td>50</td>
         </tr>
     
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2022-05-26</td>
           <td>J. Mach. Learn. Res.</td>
-          <td>11</td>
+          <td>14</td>
           <td>50</td>
         </tr>
     
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2023-11-10</td>
           <td>ArXiv</td>
-          <td>32</td>
+          <td>49</td>
           <td>6</td>
         </tr>
     
diff --git a/docs/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6.md b/docs/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6.md
index 69695770..9dcf24a0 100644
--- a/docs/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6.md
+++ b/docs/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:31 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:19 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -42,9 +42,9 @@ hide:
   </thead>
   <tbody>
     
-        <tr id="Foundation models have emerged as a promising approach in time series forecasting (TSF). Existing approaches either fine-tune large language models (LLMs) or build large-scale time-series datasets to develop TSF foundation models. However, these methods face challenges due to the severe cross-domain gap or in-domain heterogeneity. In this paper, we explore a new road to building a TSF foundation model from rich and high-quality natural images, based on the intrinsic similarities between images and time series. To bridge the gap between the two domains, we reformulate the TSF task as an image reconstruction task, which is further processed by a visual masked autoencoder (MAE) self-supervised pre-trained on the ImageNet dataset. Surprisingly, without further adaptation in the time-series domain, the proposed VisionTS could achieve superior zero-shot forecasting performance compared to existing TSF foundation models. With minimal fine-tuning, VisionTS could further improve the forecasting and achieve state-of-the-art performance in most cases. These findings suggest that visual models could be a free lunch for TSF and highlight the potential for future cross-domain research between computer vision and TSF. Our code is publicly available at https://github.com/Keytoyze/VisionTS.">
+        <tr id="Foundation models have emerged as a promising approach in time series forecasting (TSF). Existing approaches either repurpose large language models (LLMs) or build large-scale time series datasets to develop TSF foundation models for universal forecasting. However, these methods face challenges due to the severe cross-domain gap or in-domain heterogeneity. This paper explores a new road to building a TSF foundation model from rich, high-quality natural images. Our key insight is that a visual masked autoencoder, pre-trained on the ImageNet dataset, can naturally be a numeric series forecaster. By reformulating TSF as an image reconstruction task, we bridge the gap between image pre-training and TSF downstream tasks. Surprisingly, without further adaptation in the time-series domain, the proposed VisionTS could achieve superior zero-shot forecasting performance compared to existing TSF foundation models. With fine-tuning for one epoch, VisionTS could further improve the forecasting and achieve state-of-the-art performance in most cases. Extensive experiments reveal intrinsic similarities between images and real-world time series, suggesting visual models may offer a ``free lunch'' for TSF and highlight the potential for future cross-modality research. Our code is publicly available at https://github.com/Keytoyze/VisionTS.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/852d2d0acd8ab45c02a7f5fd921f76c8b5558d99" target='_blank'>VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/7665209f6d696321650c8d7eb1d1635406f44d0d" target='_blank'>VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters</a></td>
           <td>
             Mouxiang Chen, Lefei Shen, Zhuo Li, Xiaoyun Joy Wang, Jianling Sun, Chenghao Liu
           </td>
@@ -54,18 +54,6 @@ hide:
           <td>3</td>
         </tr>
     
-        <tr id="We present General Time Transformer (GTT), an encoder-only style foundation model for zero-shot multivariate time series forecasting. GTT is pretrained on a large dataset of 200M high-quality time series samples spanning diverse domains. In our proposed framework, the task of multivariate time series forecasting is formulated as a channel-wise next curve shape prediction problem, where each time series sample is represented as a sequence of non-overlapping curve shapes with a unified numerical magnitude. GTT is trained to predict the next curve shape based on a window of past curve shapes in a channel-wise manner. Experimental results demonstrate that GTT exhibits superior zero-shot multivariate forecasting capabilities on unseen time series datasets, even surpassing state-of-the-art supervised baselines. Additionally, we investigate the impact of varying GTT model parameters and training dataset scales, observing that the scaling law also holds in the context of zero-shot multivariate time series forecasting.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/15ea0190593662402d0493958a53d5ca3b8e8406" target='_blank'>Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction</a></td>
-          <td>
-            Cheng Feng, Long Huang, Denis Krompass
-          </td>
-          <td>2024-02-12</td>
-          <td>ArXiv</td>
-          <td>3</td>
-          <td>14</td>
-        </tr>
-    
         <tr id="In recent years, generative pre-trained paradigms such as Large Language Models (LLMs) and Large Vision Models (LVMs) have achieved revolutionary advancements and widespread real-world applications. Particularly, the emergence of pre-trained LLMs-based temporal works, compared to previous deep model approaches, has demonstrated superior generalization and robustness, showcasing the potential of generative pre-trained paradigms as foundation models for time series. However, those LLMs-based works mainly focus on cross-modal research, i.e., leveraging the language capabilities of LLMs in time series contexts. Although they have achieved impressive performance, there still exist the issues of concept drift caused by differences in data distribution and inflexibility caused by misalignment of dimensions. To this end, inspired by recent work on LVMs, we reconsider the paradigm of time series modeling. In this paper, we comprehensively explore, for the first time, the effectiveness and superiority of the Generative Pre-trained Diffusion (GPD) paradigm in real-world multivariate time series forecasting (TSF). Specifically, to mitigate performance bias introduced by sophisticated networks, we propose a straightforward MLP diffusion network for unconditional modeling of time series. Then we employ a zero-shot and tuning-free method to predict (generate) future data using historical data as prompts. The GPD paradigm is established on the time series modality, effectively preventing the phenomenon of concept drift, and enabling flexible forecasting of arbitrary lengths. We demonstrate that the GPD paradigm achieves comprehensive performance and generalization comparable to current SOTA LLM-based and deep model paradigms on mainstream benchmarks and various TSF tasks. Extensive experiments validate the potential of the GPD paradigm and its assistance in future related research.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/818793961396bbe4c744932676f0ff9cc8385bed" target='_blank'>Generative Pre-Trained Diffusion Paradigm for Zero-Shot Time Series Forecasting</a></td>
@@ -75,19 +63,19 @@ hide:
           <td>2024-06-04</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>4</td>
+          <td>5</td>
         </tr>
     
-        <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
+        <tr id="We present General Time Transformer (GTT), an encoder-only style foundation model for zero-shot multivariate time series forecasting. GTT is pretrained on a large dataset of 200M high-quality time series samples spanning diverse domains. In our proposed framework, the task of multivariate time series forecasting is formulated as a channel-wise next curve shape prediction problem, where each time series sample is represented as a sequence of non-overlapping curve shapes with a unified numerical magnitude. GTT is trained to predict the next curve shape based on a window of past curve shapes in a channel-wise manner. Experimental results demonstrate that GTT exhibits superior zero-shot multivariate forecasting capabilities on unseen time series datasets, even surpassing state-of-the-art supervised baselines. Additionally, we investigate the impact of varying GTT model parameters and training dataset scales, observing that the scaling law also holds in the context of zero-shot multivariate time series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4fb78450650894091b0a55b5504a8b0a6f3dec37" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/15ea0190593662402d0493958a53d5ca3b8e8406" target='_blank'>Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction</a></td>
           <td>
-            Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
+            Cheng Feng, Long Huang, Denis Krompass
           </td>
-          <td>2024-03-12</td>
+          <td>2024-02-12</td>
           <td>ArXiv</td>
-          <td>28</td>
-          <td>18</td>
+          <td>4</td>
+          <td>15</td>
         </tr>
     
         <tr id="Over the past years, foundation models have caused a paradigm shift in machine learning due to their unprecedented capabilities for zero-shot and few-shot generalization. However, despite the success of foundation models in modalities such as natural language processing and computer vision, the development of foundation models for time series forecasting has lagged behind. We present Lag-Llama, a general-purpose foundation model for univariate probabilistic time series forecasting based on a decoder-only transformer architecture that uses lags as covariates. Lag-Llama is pretrained on a large corpus of diverse time series data from several domains, and demonstrates strong zero-shot generalization capabilities compared to a wide range of forecasting models on downstream datasets across domains. Moreover, when fine-tuned on relatively small fractions of such previously unseen datasets, Lag-Llama achieves state-of-the-art performance, outperforming prior deep learning approaches, emerging as the best general-purpose model on average. Lag-Llama serves as a strong contender to the current state-of-art in time series forecasting and paves the way for future advancements in foundation models tailored to time series data.">
@@ -98,56 +86,56 @@ hide:
           </td>
           <td>2023-10-12</td>
           <td>ArXiv</td>
-          <td>16</td>
+          <td>23</td>
           <td>40</td>
         </tr>
     
-        <tr id="Time series data are ubiquitous across various domains, making time series analysis critically important. Traditional time series models are task-specific, featuring singular functionality and limited generalization capacity. Recently, large language foundation models have unveiled their remarkable capabilities for cross-task transferability, zero-shot/few-shot learning, and decision-making explainability. This success has sparked interest in the exploration of foundation models to solve multiple time series challenges simultaneously. There are two main research lines, namely pre-training foundation models from scratch for time series and adapting large language foundation models for time series. They both contribute to the development of a unified model that is highly generalizable, versatile, and comprehensible for time series analysis. This survey offers a 3E analytical framework for comprehensive examination of related research. Specifically, we examine existing works from three dimensions, namely Effectiveness, Efficiency and Explainability. In each dimension, we focus on discussing how related works devise tailored solution by considering unique challenges in the realm of time series. Furthermore, we provide a domain taxonomy to help followers keep up with the domain-specific advancements. In addition, we introduce extensive resources to facilitate the field's development, including datasets, open-source, time series libraries. A GitHub repository is also maintained for resource updates (https://github.com/start2020/Awesome-TimeSeries-LLM-FM).">
+        <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/522828823ff726598675971036fff01b57320578" target='_blank'>A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
           <td>
-            Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung
+            Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
           </td>
-          <td>2024-05-03</td>
+          <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>47</td>
+          <td>69</td>
+          <td>18</td>
         </tr>
     
-        <tr id="In this study, we present aLLM4TS, an innovative framework that adapts Large Language Models (LLMs) for time-series representation learning. Central to our approach is that we reconceive time-series forecasting as a self-supervised, multi-patch prediction task, which, compared to traditional contrastive learning or mask-and-reconstruction methods, captures temporal dynamics in patch representations more effectively. Our strategy encompasses two-stage training: (i). a causal continual pre-training phase on various time-series datasets, anchored on next patch prediction, effectively syncing LLM capabilities with the intricacies of time-series data; (ii). fine-tuning for multi-patch prediction in the targeted time-series context. A distinctive element of our framework is the patch-wise decoding layer, which departs from previous methods reliant on sequence-level decoding. Such a design directly transposes individual patches into temporal sequences, thereby significantly bolstering the model's proficiency in mastering temporal patch-based representations. aLLM4TS demonstrates superior performance in several downstream tasks, proving its effectiveness in deriving temporal representations with enhanced transferability and marking a pivotal advancement in the adaptation of LLMs for time-series analysis.">
+        <tr id="Time Series Forecasting (TSF) is key functionality in numerous fields, including in finance, weather services, and energy management. While TSF methods are emerging these days, many of them require domain-specific data collection and model training and struggle with poor generalization performance on new domains. Foundation models aim to overcome this limitation. Pre-trained on large-scale language or time series data, they exhibit promising inferencing capabilities in new or unseen data. This has spurred a surge in new TSF foundation models. We propose a new benchmark, FoundTS, to enable thorough and fair evaluation and comparison of such models. FoundTS covers a variety of TSF foundation models, including those based on large language models and those pretrained on time series. Next, FoundTS supports different forecasting strategies, including zero-shot, few-shot, and full-shot, thereby facilitating more thorough evaluations. Finally, FoundTS offers a pipeline that standardizes evaluation processes such as dataset splitting, loading, normalization, and few-shot sampling, thereby facilitating fair evaluations. Building on this, we report on an extensive evaluation of TSF foundation models on a broad range of datasets from diverse domains and with different statistical characteristics. Specifically, we identify pros and cons and inherent limitations of existing foundation models, and we identify directions for future model design. We make our code and datasets available at https://anonymous.4open.science/r/FoundTS-C2B0.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c699624c3cdd5e2ac0d1f67de733bbad1b1c0e37" target='_blank'>Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learning</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/c245a7d561855d7bef6c758e45e772129b52dacb" target='_blank'>FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting</a></td>
           <td>
-            Yuxuan Bian, Xu Ju, Jiangtong Li, Zhijian Xu, Dawei Cheng, Qiang Xu
+            Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang
           </td>
-          <td>2024-02-07</td>
+          <td>2024-10-15</td>
           <td>ArXiv</td>
-          <td>5</td>
-          <td>5</td>
+          <td>0</td>
+          <td>27</td>
         </tr>
     
-        <tr id="Time series has been left behind in the era of pre-training and transfer learning. While research in the fields of natural language processing and computer vision are enjoying progressively larger datasets to train massive models, the most popular time series datasets consist of only tens of thousands of time steps, limiting our ability to study the effectiveness of pre-training and scaling. Recent studies have also cast doubt on the need for expressive models and scale. To alleviate these issues, we introduce three large-scale time series forecasting datasets from the cloud operations (CloudOps) domain, the largest having billions of observations, enabling further study into pre-training and scaling of time series models. We build the empirical groundwork for studying pre-training and scaling of time series models and pave the way for future research by identifying a promising candidate architecture. We show that it is a strong zero-shot baseline and benefits from further scaling, both in model and dataset size. Accompanying these datasets and results is a suite of comprehensive benchmark results comparing classical and deep learning baselines to our pre-trained method - achieving a 27% reduction in error on the largest dataset. Code and datasets can be found https://github.com/SalesforceAIResearch/pretrain-time-series-cloudops.">
+        <tr id="Time series data are ubiquitous across various domains, making time series analysis critically important. Traditional time series models are task-specific, featuring singular functionality and limited generalization capacity. Recently, large language foundation models have unveiled their remarkable capabilities for cross-task transferability, zero-shot/few-shot learning, and decision-making explainability. This success has sparked interest in the exploration of foundation models to solve multiple time series challenges simultaneously. There are two main research lines, namely pre-training foundation models from scratch for time series and adapting large language foundation models for time series. They both contribute to the development of a unified model that is highly generalizable, versatile, and comprehensible for time series analysis. This survey offers a 3E analytical framework for comprehensive examination of related research. Specifically, we examine existing works from three dimensions, namely Effectiveness, Efficiency and Explainability. In each dimension, we focus on discussing how related works devise tailored solution by considering unique challenges in the realm of time series. Furthermore, we provide a domain taxonomy to help followers keep up with the domain-specific advancements. In addition, we introduce extensive resources to facilitate the field's development, including datasets, open-source, time series libraries. A GitHub repository is also maintained for resource updates (https://github.com/start2020/Awesome-TimeSeries-LLM-FM).">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/dbd6463cfff26861f6f3ea8e8d19eff4e6331904" target='_blank'>Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/522828823ff726598675971036fff01b57320578" target='_blank'>A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model</a></td>
           <td>
-            Gerald Woo, Chenghao Liu, Akshat Kumar, Doyen Sahoo
+            Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung
           </td>
-          <td>2023-10-08</td>
+          <td>2024-05-03</td>
           <td>ArXiv</td>
           <td>7</td>
-          <td>22</td>
+          <td>47</td>
         </tr>
     
-        <tr id="Despite the impressive achievements of pre-trained models in the fields of natural language processing (NLP) and computer vision (CV), progress in the domain of time series analysis has been limited. In contrast to NLP and CV, where a single model can handle various tasks, time series analysis still relies heavily on task-specific methods for activities such as classification, anomaly detection, forecasting, and few-shot learning. The primary obstacle to developing a pre-trained model for time series analysis is the scarcity of sufficient training data. In our research, we overcome this obstacle by utilizing pre-trained models from language or CV, which have been trained on billions of data points, and apply them to time series analysis. We assess the effectiveness of the pre-trained transformer model in two ways. Initially, we maintain the original structure of the self-attention and feedforward layers in the residual blocks of the pre-trained language or image model, using the Frozen Pre-trained Transformer (FPT) for time series analysis with the addition of projection matrices for input and output. Additionally, we introduce four unique adapters, designed specifically for downstream tasks based on the pre-trained model, including forecasting and anomaly detection. These adapters are further enhanced with efficient parameter tuning, resulting in superior performance compared to all state-of-the-art methods.Our comprehensive experimental studies reveal that (a) the simple FPT achieves top-tier performance across various time series analysis tasks; and (b) fine-tuning the FPT with the custom-designed adapters can further elevate its performance, outshining specialized task-specific models.">
+        <tr id="In this study, we present aLLM4TS, an innovative framework that adapts Large Language Models (LLMs) for time-series representation learning. Central to our approach is that we reconceive time-series forecasting as a self-supervised, multi-patch prediction task, which, compared to traditional contrastive learning or mask-and-reconstruction methods, captures temporal dynamics in patch representations more effectively. Our strategy encompasses two-stage training: (i). a causal continual pre-training phase on various time-series datasets, anchored on next patch prediction, effectively syncing LLM capabilities with the intricacies of time-series data; (ii). fine-tuning for multi-patch prediction in the targeted time-series context. A distinctive element of our framework is the patch-wise decoding layer, which departs from previous methods reliant on sequence-level decoding. Such a design directly transposes individual patches into temporal sequences, thereby significantly bolstering the model's proficiency in mastering temporal patch-based representations. aLLM4TS demonstrates superior performance in several downstream tasks, proving its effectiveness in deriving temporal representations with enhanced transferability and marking a pivotal advancement in the adaptation of LLMs for time-series analysis.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/15bfe54223d28c74f9e39960f7ff1d03167e285f" target='_blank'>One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/c699624c3cdd5e2ac0d1f67de733bbad1b1c0e37" target='_blank'>Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learning</a></td>
           <td>
-            Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin
+            Yuxuan Bian, Xu Ju, Jiangtong Li, Zhijian Xu, Dawei Cheng, Qiang Xu
           </td>
-          <td>2023-11-24</td>
+          <td>2024-02-07</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>7</td>
+          <td>8</td>
+          <td>8</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8.md b/docs/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8.md
index 2be78df6..72ee5f26 100644
--- a/docs/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8.md
+++ b/docs/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:07 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:53 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
     
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
     
         <tr id="Modern societies have an abundance of data yet good system models are rare. Unfortunately, many of the current system identification and machine learning techniques fail to generalize outside of the training set, producing models that violate basic physical laws. This work proposes a novel method for the Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI). SINDy-SI is an iterative method that uses Sum-of-Squares (SOS) programming to learn optimally fitted models while guaranteeing that the learned model satisfies side information, such as symmetries and physical laws. Guided by the principle of Occam's razor, that the simplest or most regularized best fitted model is typically the superior choice, during each iteration SINDy-SI prunes the basis functions associated with small coefficients, yielding a sparse dynamical model upon termination. Through several numerical experiments we will show how the combination of side information constraints and sparse polynomial representation cultivates dynamical models that obey known physical laws while displaying impressive generalized performance beyond the training set.">
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>111</td>
-          <td>65</td>
+          <td>119</td>
+          <td>67</td>
         </tr>
     
         <tr id="Sparse regression has emerged as a popular technique for learning dynamical systems from temporal data, beginning with the SINDy (Sparse Identification of Nonlinear Dynamics) framework proposed by arXiv:1509.03580. Quantifying the uncertainty inherent in differential equations learned from data remains an open problem, thus we propose leveraging recent advances in statistical inference for sparse regression to address this issue. Focusing on systems of ordinary differential equations (ODEs), SINDy assumes that each equation is a parsimonious linear combination of a few candidate functions, such as polynomials, and uses methods such as sequentially-thresholded least squares or the Lasso to identify a small subset of these functions that govern the system's dynamics. We instead employ bias-corrected versions of the Lasso and ridge regression estimators, as well as an empirical Bayes variable selection technique known as SEMMS, to estimate each ODE as a linear combination of terms that are statistically significant. We demonstrate through simulations that this approach allows us to recover the functional terms that correctly describe the dynamics more often than existing methods that do not account for uncertainty.">
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>250</td>
-          <td>12</td>
+          <td>289</td>
+          <td>13</td>
         </tr>
     
         <tr id="PySINDy is a Python package for the discovery of governing dynamical systems models from data. In particular, PySINDy provides tools for applying the sparse identification of nonlinear dynamics (SINDy) (Brunton et al. 2016) approach to model discovery. In this work we provide a brief description of the mathematical underpinnings of SINDy, an overview and demonstration of the features implemented in PySINDy (with code examples), practical advice for users, and a list of potential extensions to PySINDy. Software is available at this https URL.">
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2020-04-17</td>
           <td>arXiv: Dynamical Systems</td>
-          <td>49</td>
-          <td>65</td>
+          <td>51</td>
+          <td>67</td>
         </tr>
     
         <tr id="Data-driven modeling frameworks that adopt sparse regression techniques, such as sparse identification of nonlinear dynamics (SINDy) and its modifications, are developed to resolve difficulties in extracting underlying dynamics from experimental data. In contrast to neural-network-based methods, these methods are designed to obtain white-box analytical models. In this work, we incorporate the concept of SINDy and knowledge in the field of classical mechanics to identify interpretable and sparse expressions of total energy and the Lagrangian that shelters the hidden dynamics. Moreover, our method (hereafter referred as Lagrangian-SINDy) is developed to use knowledge of simple systems that form the system being analyzed to ensure the likelihood of correct results and to improve the learning pace. Lagrangian-SINDy is highly accurate in discovering interpretable dynamics via energy-related physical quantities. Its performance is validated with three popular multi-DOF nonlinear dynamical systems, namely the spherical pendulum, double pendulum and cart-pendulum system. Comparisons with other SINDy-based methods are made and Lagrangian-SINDy is found to provide the most compact analytical models.">
@@ -134,8 +134,8 @@ hide:
           </td>
           <td>2020-01-31</td>
           <td>IEEE Robotics and Automation Letters</td>
-          <td>23</td>
-          <td>23</td>
+          <td>24</td>
+          <td>24</td>
         </tr>
     
         <tr id="With the rapid increase of available data for complex systems, there is great interest in the extraction of physically relevant information from massive datasets. Recently, a framework called Sparse Identification of Nonlinear Dynamics (SINDy) has been introduced to identify the governing equations of dynamical systems from simulation data. In this study, we extend SINDy to stochastic dynamical systems which are frequently used to model biophysical processes. We prove the asymptotic correctness of stochastic SINDy in the infinite data limit, both in the original and projected variables. We discuss algorithms to solve the sparse regression problem arising from the practical implementation of SINDy and show that cross validation is an essential tool to determine the right level of sparsity. We demonstrate the proposed methodology on two test systems, namely, the diffusion in a one-dimensional potential and the projected dynamics of a two-dimensional diffusion process.">
@@ -146,20 +146,8 @@ hide:
           </td>
           <td>2017-12-06</td>
           <td>The Journal of chemical physics</td>
-          <td>193</td>
-          <td>45</td>
-        </tr>
-    
-        <tr id="Discovering the partial differential equations underlying spatio-temporal datasets from very limited and highly noisy observations is of paramount interest in many scientific fields. However, it remains an open question to know when model discovery algorithms based on sparse regression can actually recover the underlying physical processes. In this work, we show the design matrices used to infer the equations by sparse regression can violate the irrepresentability condition (IRC) of the Lasso, even when derived from analytical PDE solutions (i.e. without additional noise). Sparse regression techniques which can recover the true underlying model under violated IRC conditions are therefore required, leading to the introduction of the randomised adaptive Lasso. We show once the latter is integrated within the deep learning model discovery framework DeepMod, a wide variety of nonlinear and chaotic canonical PDEs can be recovered: (1) up to $\mathcal{O}(2)$ higher noise-to-sample ratios than state-of-the-art algorithms, (2) with a single set of hyperparameters, which paves the road towards truly automated model discovery.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/45633bde87d3fecdb0ef577036f2465e70c51f24" target='_blank'>Sparsistent Model Discovery</a></td>
-          <td>
-            Georges Tod, G. Both, R. Kusters
-          </td>
-          <td>2021-06-22</td>
-          <td>ArXiv</td>
-          <td>1</td>
-          <td>12</td>
+          <td>200</td>
+          <td>46</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f.md b/docs/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f.md
index 6553de83..0fe04551 100644
--- a/docs/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f.md
+++ b/docs/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:18 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:15 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,104 +50,104 @@ hide:
           </td>
           <td>2023-10-12</td>
           <td>ArXiv</td>
-          <td>16</td>
+          <td>23</td>
           <td>40</td>
         </tr>
     
-        <tr id="Large language models (LLMs) are being applied to time series tasks, particularly time series forecasting. However, are language models actually useful for time series? After a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade the forecasting results -- in most cases the results even improved. We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and reveal that patching and attention structures perform similarly to state-of-the-art LLM-based forecasters.">
+        <tr id="Motivated by the recent success of time-series foundation models for zero-shot forecasting, we present a methodology for $\textit{in-context fine-tuning}$ of a time-series foundation model. In particular, we design a pretrained foundation model that can be prompted (at inference time) with multiple time-series examples, in order to forecast a target time-series into the future. Our foundation model is specifically trained to utilize examples from multiple related time-series in its context window (in addition to the history of the target time-series) to help it adapt to the specific distribution of the target domain at inference time. We show that such a foundation model that uses in-context examples at inference time can obtain much better performance on popular forecasting benchmarks compared to supervised deep learning methods, statistical models, as well as other time-series foundation models. Interestingly, our in-context fine-tuning approach even rivals the performance of a foundation model that is explicitly fine-tuned on the target domain.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/df0d604b8e8e3b2947d9865d735f204c08635012" target='_blank'>Are Language Models Actually Useful for Time Series Forecasting?</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/1216a29c0beb36a2b621b59d36877edf05afcf8c" target='_blank'>In-Context Fine-Tuning for Time-Series Foundation Models</a></td>
           <td>
-            Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen
+            Abhimanyu Das, Matthew Faw, Rajat Sen, Yichen Zhou
           </td>
-          <td>2024-06-22</td>
+          <td>2024-10-31</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>1</td>
+          <td>0</td>
+          <td>14</td>
         </tr>
     
-        <tr id="Time series forecasting holds significant importance in many real-world dynamic systems and has been extensively studied. Unlike natural language process (NLP) and computer vision (CV), where a single large model can tackle multiple tasks, models for time series forecasting are often specialized, necessitating distinct designs for different tasks and applications. While pre-trained foundation models have made impressive strides in NLP and CV, their development in time series domains has been constrained by data sparsity. Recent studies have revealed that large language models (LLMs) possess robust pattern recognition and reasoning abilities over complex sequences of tokens. However, the challenge remains in effectively aligning the modalities of time series data and natural language to leverage these capabilities. In this work, we present Time-LLM, a reprogramming framework to repurpose LLMs for general time series forecasting with the backbone language models kept intact. We begin by reprogramming the input time series with text prototypes before feeding it into the frozen LLM to align the two modalities. To augment the LLM's ability to reason with time series data, we propose Prompt-as-Prefix (PaP), which enriches the input context and directs the transformation of reprogrammed input patches. The transformed time series patches from the LLM are finally projected to obtain the forecasts. Our comprehensive evaluations demonstrate that Time-LLM is a powerful time series learner that outperforms state-of-the-art, specialized forecasting models. Moreover, Time-LLM excels in both few-shot and zero-shot learning scenarios.">
+        <tr id="Large language models (LLMs) are being applied to time series forecasting. But are language models actually useful for time series? In a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade forecasting performance -- in most cases, the results even improve! We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and find that patching and attention structures perform similarly to LLM-based forecasters.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277" target='_blank'>Time-LLM: Time Series Forecasting by Reprogramming Large Language Models</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/df0d604b8e8e3b2947d9865d735f204c08635012" target='_blank'>Are Language Models Actually Useful for Time Series Forecasting?</a></td>
           <td>
-            Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen
+            Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen
           </td>
-          <td>2023-10-03</td>
+          <td>2024-06-22</td>
           <td>ArXiv</td>
-          <td>134</td>
-          <td>9</td>
+          <td>13</td>
+          <td>3</td>
         </tr>
     
-        <tr id="Recent Transformer-based large language models (LLMs) demonstrate in-context learning ability to perform various functions based solely on the provided context, without updating model parameters. To fully utilize the in-context capabilities in time series forecasting (TSF) problems, unlike previous Transformer-based or LLM-based time series forecasting methods, we reformulate"time series forecasting tasks"as input tokens by constructing a series of (lookback, future) pairs within the tokens. This method aligns more closely with the inherent in-context mechanisms, and is more parameter-efficient without the need of using pre-trained LLM parameters. Furthermore, it addresses issues such as overfitting in existing Transformer-based TSF models, consistently achieving better performance across full-data, few-shot, and zero-shot settings compared to previous architectures.">
+        <tr id="Inspired by recent advancements in large language models (LLMs) for Natural Language Processing (NLP), there has been a surge in research focused on developing foundational models for time series forecasting. One approach involves training LLM architectures on tokenized time series data using cross-entropy loss. Although this method has demonstrated promising results, cross-entropy loss is primarily designed for classification tasks and does not account for the distance between classes. To address this limitation, we propose using the Wasserstein loss for such architectures. To validate our approach, we fine-tuned a foundational time series model on $22$ zero-shot datasets, comparing the performance of cross-entropy loss with that of Wasserstein loss. Our results demonstrate that replacing cross-entropy loss with Wasserstein loss significantly improves point estimation.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/3777e8d6cda553cbd9e3bbd6484e613386f3a59e" target='_blank'>In-context Time Series Predictor</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/341fc8f00805b53471ae91bc2f7e1d5b4d281048" target='_blank'>Fine-Tuning a Time Series Foundation Model with Wasserstein Loss</a></td>
           <td>
-            Jiecheng Lu, Yan Sun, Shihao Yang
+            Andrei Chernov
           </td>
-          <td>2024-05-23</td>
+          <td>2024-09-18</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>0</td>
         </tr>
     
-        <tr id="Multivariate time-series forecasting is vital in various domains, e.g., economic planning and weather prediction. Deep train-from-scratch models have exhibited effective performance yet require large amounts of data, which limits real-world applicability. Recently, researchers have leveraged the representation learning transferability of pre-trained Large Language Models (LLMs) to handle limited non-linguistic datasets effectively. However, incorporating LLMs with time-series data presents challenges of limited adaptation due to different compositions between time-series and linguistic data, and the inability to process multi-scale temporal information. To tackle these challenges, we propose LLM4TS, a framework for time-series forecasting with pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the \textit{time-series alignment} stage to align LLMs with the nuances of time-series data, and the \textit{forecasting fine-tuning} stage for downstream time-series forecasting tasks. Furthermore, our framework features a novel two-level aggregation method that integrates multi-scale temporal data within pre-trained LLMs, enhancing their ability to interpret time-specific information. In experiments across 7 time-series forecasting datasets, LLM4TS is superior to existing state-of-the-art methods compared with trained-from-scratch models in full-shot scenarios, and also achieves an average improvement of 6.84% in MSE in few-shot scenarios. In addition, evaluations compared with different self-supervised learning approaches highlight LLM4TS's effectiveness with representation learning in forecasting tasks.">
+        <tr id="Time series forecasting holds significant importance in many real-world dynamic systems and has been extensively studied. Unlike natural language process (NLP) and computer vision (CV), where a single large model can tackle multiple tasks, models for time series forecasting are often specialized, necessitating distinct designs for different tasks and applications. While pre-trained foundation models have made impressive strides in NLP and CV, their development in time series domains has been constrained by data sparsity. Recent studies have revealed that large language models (LLMs) possess robust pattern recognition and reasoning abilities over complex sequences of tokens. However, the challenge remains in effectively aligning the modalities of time series data and natural language to leverage these capabilities. In this work, we present Time-LLM, a reprogramming framework to repurpose LLMs for general time series forecasting with the backbone language models kept intact. We begin by reprogramming the input time series with text prototypes before feeding it into the frozen LLM to align the two modalities. To augment the LLM's ability to reason with time series data, we propose Prompt-as-Prefix (PaP), which enriches the input context and directs the transformation of reprogrammed input patches. The transformed time series patches from the LLM are finally projected to obtain the forecasts. Our comprehensive evaluations demonstrate that Time-LLM is a powerful time series learner that outperforms state-of-the-art, specialized forecasting models. Moreover, Time-LLM excels in both few-shot and zero-shot learning scenarios.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/663dc434fabc1cf1d3e85fff3f7ddcd313035d18" target='_blank'>LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277" target='_blank'>Time-LLM: Time Series Forecasting by Reprogramming Large Language Models</a></td>
           <td>
-            Ching Chang, Wenjie Peng, Tien-Fu Chen
+            Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen
           </td>
-          <td>2023-08-16</td>
+          <td>2023-10-03</td>
           <td>ArXiv</td>
-          <td>16</td>
-          <td>2</td>
+          <td>195</td>
+          <td>9</td>
         </tr>
     
-        <tr id="Time series data are ubiquitous across various domains, making time series analysis critically important. Traditional time series models are task-specific, featuring singular functionality and limited generalization capacity. Recently, large language foundation models have unveiled their remarkable capabilities for cross-task transferability, zero-shot/few-shot learning, and decision-making explainability. This success has sparked interest in the exploration of foundation models to solve multiple time series challenges simultaneously. There are two main research lines, namely pre-training foundation models from scratch for time series and adapting large language foundation models for time series. They both contribute to the development of a unified model that is highly generalizable, versatile, and comprehensible for time series analysis. This survey offers a 3E analytical framework for comprehensive examination of related research. Specifically, we examine existing works from three dimensions, namely Effectiveness, Efficiency and Explainability. In each dimension, we focus on discussing how related works devise tailored solution by considering unique challenges in the realm of time series. Furthermore, we provide a domain taxonomy to help followers keep up with the domain-specific advancements. In addition, we introduce extensive resources to facilitate the field's development, including datasets, open-source, time series libraries. A GitHub repository is also maintained for resource updates (https://github.com/start2020/Awesome-TimeSeries-LLM-FM).">
+        <tr id="Recent Transformer-based large language models (LLMs) demonstrate in-context learning ability to perform various functions based solely on the provided context, without updating model parameters. To fully utilize the in-context capabilities in time series forecasting (TSF) problems, unlike previous Transformer-based or LLM-based time series forecasting methods, we reformulate"time series forecasting tasks"as input tokens by constructing a series of (lookback, future) pairs within the tokens. This method aligns more closely with the inherent in-context mechanisms, and is more parameter-efficient without the need of using pre-trained LLM parameters. Furthermore, it addresses issues such as overfitting in existing Transformer-based TSF models, consistently achieving better performance across full-data, few-shot, and zero-shot settings compared to previous architectures.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/522828823ff726598675971036fff01b57320578" target='_blank'>A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/3777e8d6cda553cbd9e3bbd6484e613386f3a59e" target='_blank'>In-context Time Series Predictor</a></td>
           <td>
-            Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung
+            Jiecheng Lu, Yan Sun, Shihao Yang
           </td>
-          <td>2024-05-03</td>
+          <td>2024-05-23</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>47</td>
+          <td>2</td>
+          <td>2</td>
         </tr>
     
-        <tr id="Time-series forecasting (TSF) finds broad applications in real-world scenarios. Prompting off-the-shelf Large Language Models (LLMs) demonstrates strong zero-shot TSF capabilities while preserving computational efficiency. However, existing prompting methods oversimplify TSF as language next-token predictions, overlooking its dynamic nature and lack of integration with state-of-the-art prompt strategies such as Chain-of-Thought. Thus, we propose LSTPrompt, a novel approach for prompting LLMs in zero-shot TSF tasks. LSTPrompt decomposes TSF into short-term and long-term forecasting sub-tasks, tailoring prompts to each. LSTPrompt guides LLMs to regularly reassess forecasting mechanisms to enhance adaptability. Extensive evaluations demonstrate consistently better performance of LSTPrompt than existing prompting methods, and competitive results compared to foundation TSF models.">
+        <tr id="Time Series Forecasting (TSF) is key functionality in numerous fields, including in finance, weather services, and energy management. While TSF methods are emerging these days, many of them require domain-specific data collection and model training and struggle with poor generalization performance on new domains. Foundation models aim to overcome this limitation. Pre-trained on large-scale language or time series data, they exhibit promising inferencing capabilities in new or unseen data. This has spurred a surge in new TSF foundation models. We propose a new benchmark, FoundTS, to enable thorough and fair evaluation and comparison of such models. FoundTS covers a variety of TSF foundation models, including those based on large language models and those pretrained on time series. Next, FoundTS supports different forecasting strategies, including zero-shot, few-shot, and full-shot, thereby facilitating more thorough evaluations. Finally, FoundTS offers a pipeline that standardizes evaluation processes such as dataset splitting, loading, normalization, and few-shot sampling, thereby facilitating fair evaluations. Building on this, we report on an extensive evaluation of TSF foundation models on a broad range of datasets from diverse domains and with different statistical characteristics. Specifically, we identify pros and cons and inherent limitations of existing foundation models, and we identify directions for future model design. We make our code and datasets available at https://anonymous.4open.science/r/FoundTS-C2B0.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/bbf272d92b76cc5f1e6f793dcfeb5edd4abed7ef" target='_blank'>LSTPrompt: Large Language Models as Zero-Shot Time Series Forecasters by Long-Short-Term Prompting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/c245a7d561855d7bef6c758e45e772129b52dacb" target='_blank'>FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting</a></td>
           <td>
-            Haoxin Liu, Zhiyuan Zhao, Jindong Wang, Harshavardhan Kamarthi, B. A. Prakash
+            Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang
           </td>
-          <td>2024-02-25</td>
+          <td>2024-10-15</td>
           <td>ArXiv</td>
-          <td>6</td>
-          <td>7</td>
+          <td>0</td>
+          <td>27</td>
         </tr>
     
-        <tr id="Unlike natural language processing and computer vision, the development of Foundation Models (FMs) for time series forecasting is blocked due to data scarcity. While recent efforts are focused on building such FMs by unlocking the potential of language models (LMs) for time series analysis, dedicated parameters for various downstream forecasting tasks need training, which hinders the common knowledge sharing across domains. Moreover, data owners may hesitate to share the access to local data due to privacy concerns and copyright protection, which makes it impossible to simply construct a FM on cross-domain training instances. To address these issues, we propose Time-FFM, a Federated Foundation Model for Time series forecasting by leveraging pretrained LMs. Specifically, we begin by transforming time series into the modality of text tokens. To bootstrap LMs for time series reasoning, we propose a prompt adaption module to determine domain-customized prompts dynamically instead of artificially. Given the data heterogeneity across domains, we design a personalized federated training strategy by learning global encoders and local prediction heads. Our comprehensive experiments indicate that Time-FFM outperforms state-of-the-arts and promises effective few-shot and zero-shot forecaster.">
+        <tr id="Multivariate time-series forecasting is vital in various domains, e.g., economic planning and weather prediction. Deep train-from-scratch models have exhibited effective performance yet require large amounts of data, which limits real-world applicability. Recently, researchers have leveraged the representation learning transferability of pre-trained Large Language Models (LLMs) to handle limited non-linguistic datasets effectively. However, incorporating LLMs with time-series data presents challenges of limited adaptation due to different compositions between time-series and linguistic data, and the inability to process multi-scale temporal information. To tackle these challenges, we propose LLM4TS, a framework for time-series forecasting with pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the \textit{time-series alignment} stage to align LLMs with the nuances of time-series data, and the \textit{forecasting fine-tuning} stage for downstream time-series forecasting tasks. Furthermore, our framework features a novel two-level aggregation method that integrates multi-scale temporal data within pre-trained LLMs, enhancing their ability to interpret time-specific information. In experiments across 7 time-series forecasting datasets, LLM4TS is superior to existing state-of-the-art methods compared with trained-from-scratch models in full-shot scenarios, and also achieves an average improvement of 6.84% in MSE in few-shot scenarios. In addition, evaluations compared with different self-supervised learning approaches highlight LLM4TS's effectiveness with representation learning in forecasting tasks.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/68a65925ac6d270ca27a4fe99e58cf2ed0795821" target='_blank'>Time-FFM: Towards LM-Empowered Federated Foundation Model for Time Series Forecasting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/663dc434fabc1cf1d3e85fff3f7ddcd313035d18" target='_blank'>LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters</a></td>
           <td>
-            Qingxiang Liu, Xu Liu, Chenghao Liu, Qingsong Wen, Yuxuan Liang
+            Ching Chang, Wenjie Peng, Tien-Fu Chen
           </td>
-          <td>2024-05-23</td>
+          <td>2023-08-16</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>6</td>
+          <td>22</td>
+          <td>3</td>
         </tr>
     
-        <tr id="Despite the impressive achievements of pre-trained models in the fields of natural language processing (NLP) and computer vision (CV), progress in the domain of time series analysis has been limited. In contrast to NLP and CV, where a single model can handle various tasks, time series analysis still relies heavily on task-specific methods for activities such as classification, anomaly detection, forecasting, and few-shot learning. The primary obstacle to developing a pre-trained model for time series analysis is the scarcity of sufficient training data. In our research, we overcome this obstacle by utilizing pre-trained models from language or CV, which have been trained on billions of data points, and apply them to time series analysis. We assess the effectiveness of the pre-trained transformer model in two ways. Initially, we maintain the original structure of the self-attention and feedforward layers in the residual blocks of the pre-trained language or image model, using the Frozen Pre-trained Transformer (FPT) for time series analysis with the addition of projection matrices for input and output. Additionally, we introduce four unique adapters, designed specifically for downstream tasks based on the pre-trained model, including forecasting and anomaly detection. These adapters are further enhanced with efficient parameter tuning, resulting in superior performance compared to all state-of-the-art methods.Our comprehensive experimental studies reveal that (a) the simple FPT achieves top-tier performance across various time series analysis tasks; and (b) fine-tuning the FPT with the custom-designed adapters can further elevate its performance, outshining specialized task-specific models.">
+        <tr id="Deep learning for time series forecasting has seen significant advancements over the past decades. However, despite the success of large-scale pre-training in language and vision domains, pre-trained time series models remain limited in scale and operate at a high cost, hindering the development of larger capable forecasting models in real-world applications. In response, we introduce Time-MoE, a scalable and unified architecture designed to pre-train larger, more capable forecasting foundation models while reducing inference costs. By leveraging a sparse mixture-of-experts (MoE) design, Time-MoE enhances computational efficiency by activating only a subset of networks for each prediction, reducing computational load while maintaining high model capacity. This allows Time-MoE to scale effectively without a corresponding increase in inference costs. Time-MoE comprises a family of decoder-only transformer models that operate in an auto-regressive manner and support flexible forecasting horizons with varying input context lengths. We pre-trained these models on our newly introduced large-scale data Time-300B, which spans over 9 domains and encompassing over 300 billion time points. For the first time, we scaled a time series foundation model up to 2.4 billion parameters, achieving significantly improved forecasting precision. Our results validate the applicability of scaling laws for training tokens and model size in the context of time series forecasting. Compared to dense models with the same number of activated parameters or equivalent computation budgets, our models consistently outperform them by large margin. These advancements position Time-MoE as a state-of-the-art solution for tackling real-world time series forecasting challenges with superior capability, efficiency, and flexibility.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/15bfe54223d28c74f9e39960f7ff1d03167e285f" target='_blank'>One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/2cb3044ef42c7ee022a988864028b80ce977072c" target='_blank'>Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts</a></td>
           <td>
-            Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin
+            X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin
           </td>
-          <td>2023-11-24</td>
+          <td>2024-09-24</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>7</td>
+          <td>2</td>
+          <td>9</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90.md b/docs/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90.md
index 1f7696d5..6d974095 100644
--- a/docs/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90.md
+++ b/docs/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90.md
@@ -11,7 +11,7 @@ hide:
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:34 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:30 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,20 @@ hide:
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>4</td>
-          <td>66</td>
+          <td>21</td>
+          <td>67</td>
+        </tr>
+    
+        <tr id="In natural language processing and computer vision, self-supervised pre-training on large datasets unlocks foundational model capabilities across domains and tasks. However, this potential has not yet been realised in time series analysis, where existing methods disregard the heterogeneous nature of time series characteristics. Time series are prevalent in many domains, including medicine, engineering, natural sciences, and finance, but their characteristics vary significantly in terms of variate count, inter-variate relationships, temporal dynamics, and sampling frequency. This inherent heterogeneity across domains prevents effective pre-training on large time series corpora. To address this issue, we introduce OTiS, an open model for general time series analysis, that has been specifically designed to handle multi-domain heterogeneity. We propose a novel pre-training paradigm including a tokeniser with learnable domain-specific signatures, a dual masking strategy to capture temporal causality, and a normalised cross-correlation loss to model long-range dependencies. Our model is pre-trained on a large corpus of 640,187 samples and 11 billion time points spanning 8 distinct domains, enabling it to analyse time series from any (unseen) domain. In comprehensive experiments across 15 diverse applications - including classification, regression, and forecasting - OTiS showcases its ability to accurately capture domain-specific data characteristics and demonstrates its competitiveness against state-of-the-art baselines. Our code and pre-trained weights are publicly available at https://github.com/oetu/otis.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/67381675766ffe648e1346cb19d0d7318366bbbe" target='_blank'>Towards Generalisable Time Series Understanding Across Domains</a></td>
+          <td>
+            Ozgun Turgut, Philip Muller, M. Menten, Daniel Rueckert
+          </td>
+          <td>2024-10-09</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>15</td>
         </tr>
     
         <tr id="Large pre-trained models have been vital in recent advancements in domains like language and vision, making model training for individual downstream tasks more efficient and provide superior performance. However, tackling time-series analysis tasks usually involves designing and training a separate model from scratch leveraging training data and domain expertise specific to the task. We tackle a significant challenge for pre-training a foundational time-series model from multi-domain time-series datasets: extracting semantically useful tokenized inputs to the model across heterogenous time-series from different domains. We propose Large Pre-trained Time-series Models (LPTM) that introduces a novel method of \textit{adaptive segmentation} that automatically identifies optimal dataset-specific segmentation strategy during pre-training. This enables LPTM to perform similar to or better than domain-specific state-of-art model when fine-tuned to different downstream time-series analysis tasks and under zero-shot settings. LPTM achieves superior forecasting and time-series classification results taking up to 40% less data and 50% less training time compared to state-of-art baselines.">
@@ -62,8 +74,8 @@ hide:
           </td>
           <td>2023-11-19</td>
           <td>ArXiv</td>
-          <td>2</td>
-          <td>7</td>
+          <td>4</td>
+          <td>8</td>
         </tr>
     
         <tr id="The field of general time series analysis has recently begun to explore unified modeling, where a common architectural backbone can be retrained on a specific task for a specific dataset. In this work, we approach unification from a complementary vantage point: unification across tasks and domains. To this end, we explore the impact of discrete, learnt, time series data representations that enable generalist, cross-domain training. Our method, TOTEM, or TOkenized Time Series EMbeddings, proposes a simple tokenizer architecture that embeds time series data from varying domains using a discrete vectorized representation learned in a self-supervised manner. TOTEM works across multiple tasks and domains with minimal to no tuning. We study the efficacy of TOTEM with an extensive evaluation on 17 real world time series datasets across 3 tasks. We evaluate both the specialist (i.e., training a model on each domain) and generalist (i.e., training a single model on many domains) settings, and show that TOTEM matches or outperforms previous best methods on several popular benchmarks. The code can be found at: https://github.com/SaberaTalukder/TOTEM.">
@@ -74,7 +86,7 @@ hide:
           </td>
           <td>2024-02-26</td>
           <td>ArXiv</td>
-          <td>4</td>
+          <td>7</td>
           <td>4</td>
         </tr>
     
@@ -86,8 +98,8 @@ hide:
           </td>
           <td>2024-01-08</td>
           <td>ArXiv</td>
-          <td>5</td>
           <td>6</td>
+          <td>7</td>
         </tr>
     
         <tr id="Despite the impressive achievements of pre-trained models in the fields of natural language processing (NLP) and computer vision (CV), progress in the domain of time series analysis has been limited. In contrast to NLP and CV, where a single model can handle various tasks, time series analysis still relies heavily on task-specific methods for activities such as classification, anomaly detection, forecasting, and few-shot learning. The primary obstacle to developing a pre-trained model for time series analysis is the scarcity of sufficient training data. In our research, we overcome this obstacle by utilizing pre-trained models from language or CV, which have been trained on billions of data points, and apply them to time series analysis. We assess the effectiveness of the pre-trained transformer model in two ways. Initially, we maintain the original structure of the self-attention and feedforward layers in the residual blocks of the pre-trained language or image model, using the Frozen Pre-trained Transformer (FPT) for time series analysis with the addition of projection matrices for input and output. Additionally, we introduce four unique adapters, designed specifically for downstream tasks based on the pre-trained model, including forecasting and anomaly detection. These adapters are further enhanced with efficient parameter tuning, resulting in superior performance compared to all state-of-the-art methods.Our comprehensive experimental studies reveal that (a) the simple FPT achieves top-tier performance across various time series analysis tasks; and (b) fine-tuning the FPT with the custom-designed adapters can further elevate its performance, outshining specialized task-specific models.">
@@ -99,18 +111,18 @@ hide:
           <td>2023-11-24</td>
           <td>ArXiv</td>
           <td>4</td>
-          <td>7</td>
+          <td>8</td>
         </tr>
     
         <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4fb78450650894091b0a55b5504a8b0a6f3dec37" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
           <td>
             Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>28</td>
+          <td>69</td>
           <td>18</td>
         </tr>
     
@@ -122,8 +134,8 @@ hide:
           </td>
           <td>2024-02-04</td>
           <td>ArXiv</td>
-          <td>32</td>
-          <td>22</td>
+          <td>71</td>
+          <td>25</td>
         </tr>
     
   </tbody>
diff --git a/site/Symbolic regression/index.html b/site/Symbolic regression/index.html
index 78195032..bad5505f 100644
--- a/site/Symbolic regression/index.html	
+++ b/site/Symbolic regression/index.html	
@@ -451,7 +451,7 @@ <h1>Symbolic regression</h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:18 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:06:01 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -515,9 +515,9 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
             S. Brunton, J. Proctor, J. Kutz
           </td>
           <td>2015-09-11</td>
-          <td>Proceedings of the National Academy of Sciences, Proceedings of the National Academy of Sciences of the United States of America</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>Proceedings of the National Academy of Sciences of the United States of America, Proceedings of the National Academy of Sciences</td>
+          <td>3408</td>
+          <td>67</td>
 
             <td><a href='../recommendations/5d150cec2775f9bc863760448f14104cc8f42368' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -536,8 +536,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2021-02-24</td>
           <td>Nature Communications</td>
-          <td>86</td>
-          <td>23</td>
+          <td>91</td>
+          <td>24</td>
 
             <td><a href='../recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -556,8 +556,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2019-03-29</td>
           <td>Proceedings of the National Academy of Sciences of the United States of America</td>
-          <td>615</td>
-          <td>65</td>
+          <td>648</td>
+          <td>67</td>
 
             <td><a href='../recommendations/3c9961153493370500020c81527b3548c96f81e0' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -576,8 +576,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2016-08-18</td>
           <td>Nature Communications</td>
-          <td>451</td>
-          <td>65</td>
+          <td>468</td>
+          <td>67</td>
 
             <td><a href='../recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -595,9 +595,9 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
             E. Kaiser, J. Kutz, S. Brunton
           </td>
           <td>2017-11-15</td>
-          <td>Proceedings. Mathematical, Physical, and Engineering Sciences, Proceedings of the Royal Society A</td>
-          <td>435</td>
-          <td>65</td>
+          <td>Proceedings of the Royal Society A, Proceedings. Mathematical, Physical, and Engineering Sciences</td>
+          <td>461</td>
+          <td>67</td>
 
             <td><a href='../recommendations/b2eb064f432557c59ce99834d7dc7817e4687271' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -615,9 +615,9 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
             N. Mangan, S. Brunton, J. Proctor, J. Kutz
           </td>
           <td>2016-05-26</td>
-          <td>IEEE Transactions on Molecular Biological and Multi-Scale Communications, IEEE Transactions on Molecular, Biological and Multi-Scale Communications</td>
-          <td>321</td>
-          <td>65</td>
+          <td>IEEE Transactions on Molecular, Biological and Multi-Scale Communications, IEEE Transactions on Molecular Biological and Multi-Scale Communications</td>
+          <td>334</td>
+          <td>67</td>
 
             <td><a href='../recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -635,9 +635,9 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
             Kadierdan Kaheman, J. Kutz, S. Brunton
           </td>
           <td>2020-04-05</td>
-          <td>Proceedings. Mathematical, Physical, and Engineering Sciences, Proceedings of the Royal Society A</td>
-          <td>202</td>
-          <td>65</td>
+          <td>Proceedings of the Royal Society A, Proceedings. Mathematical, Physical, and Engineering Sciences</td>
+          <td>221</td>
+          <td>67</td>
 
             <td><a href='../recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -656,8 +656,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2018-09-07</td>
           <td>Journal of Computational and Nonlinear Dynamics</td>
-          <td>62</td>
-          <td>77</td>
+          <td>65</td>
+          <td>79</td>
 
             <td><a href='../recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -676,8 +676,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2019-09-18</td>
           <td>ArXiv, arXiv.org</td>
-          <td>32</td>
-          <td>65</td>
+          <td>35</td>
+          <td>67</td>
 
             <td><a href='../recommendations/73dd9c49f205280991826b2ea4b50344203916b4' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -696,8 +696,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2019-06-19</td>
           <td>Frontiers in Artificial Intelligence</td>
-          <td>71</td>
-          <td>65</td>
+          <td>77</td>
+          <td>67</td>
 
             <td><a href='../recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -716,8 +716,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2016-09-21</td>
           <td>Science Advances</td>
-          <td>1187</td>
-          <td>65</td>
+          <td>1235</td>
+          <td>67</td>
 
             <td><a href='../recommendations/0acd117521ef5aafb09fed02ab415523b330b058' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -735,9 +735,9 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
             Urban Fasel, J. Kutz, Bingni W. Brunton, S. Brunton
           </td>
           <td>2021-11-22</td>
-          <td>Proceedings. Mathematical, Physical, and Engineering Sciences, Proceedings of the Royal Society A</td>
-          <td>161</td>
-          <td>65</td>
+          <td>Proceedings of the Royal Society A, Proceedings. Mathematical, Physical, and Engineering Sciences</td>
+          <td>181</td>
+          <td>67</td>
 
             <td><a href='../recommendations/883547fdbd88552328a6615ec620f96e39c57018' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -756,8 +756,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2022-06-01</td>
           <td>Nonlinear Dynamics</td>
-          <td>29</td>
-          <td>91</td>
+          <td>30</td>
+          <td>92</td>
 
             <td><a href='../recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -776,8 +776,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Symbolic regr
           </td>
           <td>2018-07-14</td>
           <td>IEEE Access</td>
-          <td>116</td>
-          <td>65</td>
+          <td>124</td>
+          <td>67</td>
 
             <td><a href='../recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -820,694 +820,724 @@ <h3 id="recommended_articles">Recommended articles on <i>Symbolic regression</i>
     </thead>
     <tbody>
 
-        <tr id="Over the past few years, equation discovery has gained popularity in different fields of science and engineering. However, existing equation discovery algorithms rely on the availability of noisy measurements of the state variables (i.e., displacement {and velocity}). This is a major bottleneck in structural dynamics, where we often only have access to acceleration measurements. To that end, this paper introduces a novel equation discovery algorithm for discovering governing equations of dynamical systems from acceleration-only measurements. The proposed algorithm employs a library-based approach for equation discovery. To enable equation discovery from acceleration-only measurements, we propose a novel Approximate Bayesian Computation (ABC) model that prioritizes parsimonious models. The efficacy of the proposed algorithm is illustrated using {four} structural dynamics examples that include both linear and nonlinear dynamical systems. The case studies presented illustrate the possible application of the proposed approach for equation discovery of dynamical systems from acceleration-only measurements.">
+        <tr id="Sparse Identification of Nonlinear Dynamical Systems (SINDy) is a powerful tool for the data-driven discovery of governing equations. However, it encounters challenges when modeling complex dynamical systems involving high-order derivatives or discontinuities, particularly in the presence of noise. These limitations restrict its applicability across various fields in applied mathematics and physics. To mitigate these, we propose Laplace-Enhanced SparSe Identification of Nonlinear Dynamical Systems (LES-SINDy). By transforming time-series measurements from the time domain to the Laplace domain using the Laplace transform and integration by parts, LES-SINDy enables more accurate approximations of derivatives and discontinuous terms. It also effectively handles unbounded growth functions and accumulated numerical errors in the Laplace domain, thereby overcoming challenges in the identification process. The model evaluation process selects the most accurate and parsimonious dynamical systems from multiple candidates. Experimental results across diverse ordinary and partial differential equations show that LES-SINDy achieves superior robustness, accuracy, and parsimony compared to existing methods.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/f046e6be05e115bc55624ef726d92d6d06303346" target='_blank'>
-              Discovering governing equation in structural dynamics from acceleration-only measurements
+          <td><a href="https://www.semanticscholar.org/paper/cc9d7f1e9a7f7b37237e52d46d21299f816a377c" target='_blank'>
+              LES-SINDy: Laplace-Enhanced Sparse Identification of Nonlinear Dynamical Systems
               </a>
             </td>
           <td>
-            Calvin Alvares, Souvik Chakraborty
+            Haoyang Zheng, Guang Lin
           </td>
-          <td>2024-07-18</td>
+          <td>2024-11-04</td>
           <td>ArXiv</td>
           <td>0</td>
           <td>1</td>
         </tr>
 
-        <tr id="The combination of machine learning (ML) and sparsity-promoting techniques is enabling direct extraction of governing equations from data, revolutionizing computational modeling in diverse fields of science and engineering. The discovered dynamical models could be used to address challenges in climate science, neuroscience, ecology, finance, epidemiology, and beyond. However, most existing sparse identification methods for discovering dynamical systems treat the whole system as one without considering the interactions between subsystems. As a result, such models are not able to capture small changes in the emergent system behavior. To address this issue, we developed a new method called Sparse Identification of Nonlinear Dynamical Systems from Graph-structured data (SINDyG), which incorporates the network structure into sparse regression to identify model parameters that explain the underlying network dynamics. SINDyG discovers the governing equations of network dynamics while offering improvements in accuracy and model simplicity.">
+        <tr id="Identifying dynamical systems characterized by nonlinear parameters presents significant challenges in deriving mathematical models that enhance understanding of physics. Traditional methods, such as Sparse Identification of Nonlinear Dynamics (SINDy) and symbolic regression, can extract governing equations from observational data; however, they also come with distinct advantages and disadvantages. This paper introduces a novel method within the SINDy framework, termed ADAM-SINDy, which synthesizes the strengths of established approaches by employing the ADAM optimization algorithm. This facilitates the simultaneous optimization of nonlinear parameters and coefficients associated with nonlinear candidate functions, enabling precise parameter estimation without requiring prior knowledge of nonlinear characteristics such as trigonometric frequencies, exponential bandwidths, or polynomial exponents, thereby addressing a key limitation of SINDy. Through an integrated global optimization, ADAM-SINDy dynamically adjusts all unknown variables in response to data, resulting in an adaptive identification procedure that reduces the sensitivity to the library of candidate functions. The performance of the ADAM-SINDy methodology is demonstrated across a spectrum of dynamical systems, including benchmark coupled nonlinear ordinary differential equations such as oscillators, chaotic fluid flows, reaction kinetics, pharmacokinetics, as well as nonlinear partial differential equations (wildfire transport). The results demonstrate significant improvements in identifying parameterized dynamical systems and underscore the importance of concurrently optimizing all parameters, particularly those characterized by nonlinear parameters. These findings highlight the potential of ADAM-SINDy to extend the applicability of the SINDy framework in addressing more complex challenges in dynamical system identification.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4da518ae886702aed27abdd8f94fea3042ab9377" target='_blank'>
-              Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems
+          <td><a href="https://www.semanticscholar.org/paper/0c327ed253a2c7b658940860bcf0a9e077d634b6" target='_blank'>
+              ADAM-SINDy: An Efficient Optimization Framework for Parameterized Nonlinear Dynamical System Identification
               </a>
             </td>
           <td>
-            Mohammad Amin Basiri, Sina Khanmohammadi
+            Siva Viknesh, Younes Tatari, Amirhossein Arzani
           </td>
-          <td>2024-09-02</td>
+          <td>2024-10-21</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>3</td>
+          <td>24</td>
         </tr>
 
-        <tr id="This work presents a data-driven method for learning low-dimensional time-dependent physics-based surrogate models whose predictions are endowed with uncertainty estimates. We use the operator inference approach to model reduction that poses the problem of learning low-dimensional model terms as a regression of state space data and corresponding time derivatives by minimizing the residual of reduced system equations. Standard operator inference models perform well with accurate training data that are dense in time, but producing stable and accurate models when the state data are noisy and/or sparse in time remains a challenge. Another challenge is the lack of uncertainty estimation for the predictions from the operator inference models. Our approach addresses these challenges by incorporating Gaussian process surrogates into the operator inference framework to (1) probabilistically describe uncertainties in the state predictions and (2) procure analytical time derivative estimates with quantified uncertainties. The formulation leads to a generalized least-squares regression and, ultimately, reduced-order models that are described probabilistically with a closed-form expression for the posterior distribution of the operators. The resulting probabilistic surrogate model propagates uncertainties from the observed state data to reduced-order predictions. We demonstrate the method is effective for constructing low-dimensional models of two nonlinear partial differential equations representing a compressible flow and a nonlinear diffusion-reaction process, as well as for estimating the parameters of a low-dimensional system of nonlinear ordinary differential equations representing compartmental models in epidemiology.">
+        <tr id="A significant challenge in many fields of science and engineering is making sense of time-dependent measurement data by recovering governing equations in the form of differential equations. We focus on finding parsimonious ordinary differential equation (ODE) models for nonlinear, noisy, and non-autonomous dynamical systems and propose a machine learning method for data-driven system identification. While many methods tackle noisy and limited data, non-stationarity - where differential equation parameters change over time - has received less attention. Our method, dynamic SINDy, combines variational inference with SINDy (sparse identification of nonlinear dynamics) to model time-varying coefficients of sparse ODEs. This framework allows for uncertainty quantification of ODE coefficients, expanding on previous methods for autonomous systems. These coefficients are then interpreted as latent variables and added to the system to obtain an autonomous dynamical model. We validate our approach using synthetic data, including nonlinear oscillators and the Lorenz system, and apply it to neuronal activity data from C. elegans. Dynamic SINDy uncovers a global nonlinear model, showing it can handle real, noisy, and chaotic datasets. We aim to apply our method to a variety of problems, specifically dynamic systems with complex time-dependent parameters.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/6c14dacd4e17965fe6d0db5f5cefb7cb8547dbb3" target='_blank'>
-              Bayesian learning with Gaussian processes for low-dimensional representations of time-dependent nonlinear systems
+          <td><a href="https://www.semanticscholar.org/paper/09cc57056fc3b8128466bf0ef72e8a8c737c9c51" target='_blank'>
+              Deep Generative Modeling for Identification of Noisy, Non-Stationary Dynamical Systems
               </a>
             </td>
           <td>
-            Shane A. McQuarrie, Anirban Chaudhuri, Karen Willcox, Mengwu Guo
+            Doris Voina, S. Brunton, J. Kutz
           </td>
-          <td>2024-08-06</td>
+          <td>2024-10-02</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>6</td>
+          <td>67</td>
         </tr>
 
-        <tr id="The Sparse Identification of Nonlinear Dynamics (SINDy) framework is a robust method for identifying governing equations, successfully applied to ordinary, partial, and stochastic differential equations. In this work we extend SINDy to identify delay differential equations by using an augmented library that includes delayed samples and Bayesian optimization. To identify a possibly unknown delay we minimize the reconstruction error over a set of candidates. The resulting methodology improves the overall performance by remarkably reducing the number of calls to SINDy with respect to a brute force approach. We also address a multivariate setting to identify multiple unknown delays and (non-multiplicative) parameters. Several numerical tests on delay differential equations with different long-term behavior, number of variables, delays, and parameters support the use of Bayesian optimization highlighting both the efficacy of the proposed methodology and its computational advantages. As a consequence, the class of discoverable models is significantly expanded.">
+        <tr id="Identifying nonlinear dynamics and characterizing noise from data is critical across science and engineering for understanding and modeling the behavior of the systems accurately. The modified sparse identification of nonlinear dynamics (mSINDy) has emerged as an effective framework for identifying systems embedded in heavy noise; however, further improvements can expand its capabilities and robustness. By integrating the weak SINDy (WSINDy) into mSINDy, we introduce the weak mSINDy (WmSINDy) to improve the system identification and noise modeling by harnessing the strengths of both approaches. The proposed algorithm simultaneously identifies parsimonious nonlinear dynamics and extracts noise probability distributions using automatic differentiation. We evaluate WmSINDy using several nonlinear systems and it demonstrates improved accuracy and noise characterization over baselines for systems embedded in relatively strong noise.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/827638609aa7a84d10096156bc7ead43d8060c5a" target='_blank'>
-              Data-driven Discovery of Delay Differential Equations with Discrete Delays
+          <td><a href="https://www.semanticscholar.org/paper/55056424073ed91375d017c96fd9e74bb1492e1a" target='_blank'>
+              Weak-form modified sparse identification of nonlinear dynamics
               </a>
             </td>
           <td>
-            Alessandro Pecile, N. Demo, M. Tezzele, G. Rozza, Dimitri Breda
+            Cristian L'opez, 'Angel Naranjo, Diego Salazar, Keegan J. Moore
           </td>
-          <td>2024-07-29</td>
+          <td>2024-10-23</td>
           <td>ArXiv</td>
+          <td>0</td>
           <td>1</td>
-          <td>50</td>
         </tr>
 
-        <tr id="We present a computational technique for modeling the evolution of dynamical systems in a reduced basis, with a focus on the challenging problem of modeling partially-observed partial differential equations (PDEs) on high-dimensional non-uniform grids. We address limitations of previous work on data-driven flow map learning in the sense that we focus on noisy and limited data to move toward data collection scenarios in real-world applications. Leveraging recent work on modeling PDEs in modal and nodal spaces, we present a neural network structure that is suitable for PDE modeling with noisy and limited data available only on a subset of the state variables or computational domain. In particular, spatial grid-point measurements are reduced using a learned linear transformation, after which the dynamics are learned in this reduced basis before being transformed back out to the nodal space. This approach yields a drastically reduced parameterization of the neural network compared with previous flow map models for nodal space learning. This primarily allows for smaller training data sets, but also enables reduced training times.">
+        <tr id="Biological systems exhibit complex dynamics that differential equations can often adeptly represent. Ordinary differential equation models are widespread; until recently their construction has required extensive prior knowledge of the system. Machine learning methods offer alternative means of model construction: differential equation models can be learnt from data via model discovery using sparse identification of nonlinear dynamics (SINDy). However, SINDy struggles with realistic levels of biological noise and does not incorporate prior knowledge of the system. We propose a data-driven framework for model discovery and model selection using hybrid dynamical systems: partial models containing missing terms. Neural networks are used to approximate the unknown dynamics of a system, enabling the denoising the data while simultaneously learning the latent dynamics. Simulations from the fitted neural network are then used to infer models using SINDy. We show, via model selection, that model discovery in SINDy with hybrid dynamical systems outperforms alternative approaches. We find it possible to infer models correctly up to high levels of biological noise of different types. We demonstrate the potential to learn models from sparse, noisy data in application to a canonical cell state transition using data derived from single-cell transcriptomics. Overall, this approach provides a practical framework for model discovery in biology in cases where data are noisy and sparse, of particular utility when the underlying biological mechanisms are partially but incompletely known.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/33821b7c0acd5258ca0c60a09ffdc55439ac7ac2" target='_blank'>
-              Principal Component Flow Map Learning of PDEs from Incomplete, Limited, and Noisy Data
+          <td><a href="https://www.semanticscholar.org/paper/596f63cbf87cf14a0e7859c17058a82b15c760ef" target='_blank'>
+              Data-driven model discovery and model selection for noisy biological systems
               </a>
             </td>
           <td>
-            Victor Churchill
+            Xiaojun Wu, MeiLu McDermott, Adam L. Maclean
           </td>
-          <td>2024-07-15</td>
-          <td>ArXiv</td>
-          <td>0</td>
+          <td>2024-10-04</td>
+          <td>bioRxiv</td>
           <td>0</td>
+          <td>19</td>
         </tr>
 
-        <tr id="Model parsimony is an important \emph{cognitive bias} in data-driven modelling that aids interpretability and helps to prevent over-fitting. Sparse identification of nonlinear dynamics (SINDy) methods are able to learn sparse representations of complex dynamics directly from data, given a basis of library functions. In this work, a novel Bayesian treatment of dictionary learning system identification, as an alternative to SINDy, is envisaged. The proposed method -- Bayesian identification of nonlinear dynamics (BINDy) -- is distinct from previous approaches in that it targets the full joint posterior distribution over both the terms in the library and their parameterisation in the model. This formulation confers the advantage that an arbitrary prior may be placed over the model structure to produce models that are sparse in the model space rather than in parameter space. Because this posterior is defined over parameter vectors that can change in dimension, the inference cannot be performed by standard techniques. Instead, a Gibbs sampler based on reversible-jump Markov-chain Monte-Carlo is proposed. BINDy is shown to compare favourably to ensemble SINDy in three benchmark case-studies. In particular, it is seen that the proposed method is better able to assign high probability to correct model terms.">
+        <tr id="Sparse system identification of nonlinear dynamic systems is still challenging, especially for stiff and high-order differential equations for noisy measurement data. The use of highly correlated functions makes distinguishing between true and false functions difficult, which limits the choice of functions. In this study, an equation discovery method has been proposed to tackle these problems. The key elements include a) use of B-splines for data fitting to get analytical derivatives superior to numerical derivatives, b) sequentially regularized derivatives for denoising (SRDD) algorithm, highly effective in removing noise from signal without system information loss, c) uncorrelated component analysis (UCA) algorithm that identifies and eliminates highly correlated functions while retaining the true functions, and d) physics-informed spline fitting (PISF) where the spline fitting is updated gradually while satisfying the governing equation with a dictionary of candidate functions to converge to the correct equation sequentially. The complete framework is built on a unified deep-learning architecture that eases the optimization process. The proposed method is demonstrated to discover various differential equations at various noise levels, including three-dimensional, fourth-order, and stiff equations. The parameter estimation converges accurately to the true values with a small coefficient of variation, suggesting robustness to the noise.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/04666a9e47b5508a0c0f966265fe51da2b2c634a" target='_blank'>
-              BINDy -- Bayesian identification of nonlinear dynamics with reversible-jump Markov-chain Monte-Carlo
+          <td><a href="https://www.semanticscholar.org/paper/e1e652817a3da4ffa0d9ce7fcc419c996835b61c" target='_blank'>
+              Physics-informed AI and ML-based sparse system identification algorithm for discovery of PDE's representing nonlinear dynamic systems
               </a>
             </td>
           <td>
-            M.D. Champneys, T. J. Rogers
+            A. Pal, Sutanu Bhowmick, Satish Nagarajaiah
           </td>
-          <td>2024-08-15</td>
+          <td>2024-10-13</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>9</td>
         </tr>
 
-        <tr id="Accurate and efficient fluid flow models are essential for applications relating to many physical phenomena including geophysical, aerodynamic, and biological systems. While these flows may exhibit rich and multiscale dynamics, in many cases underlying low-rank structures exist which describe the bulk of the motion. These structures tend to be spatially large and temporally slow, and may contain most of the energy in a given flow. The extraction and parsimonious representation of these low-rank dynamics from high-dimensional data is a key challenge. Inspired by the success of physics-informed machine learning methods, we propose a spectrally-informed approach to extract low-rank models of fluid flows by leveraging known spectral properties in the learning process. We incorporate this knowledge by imposing regularizations on the learned dynamics, which bias the training process towards learning low-frequency structures with corresponding higher power. We demonstrate the effectiveness of this method to improve prediction and produce learned models which better match the underlying spectral properties of prototypical fluid flows.">
+        <tr id="Effectively controlling systems governed by Partial Differential Equations (PDEs) is crucial in several fields of Applied Sciences and Engineering. These systems usually yield significant challenges to conventional control schemes due to their nonlinear dynamics, partial observability, high-dimensionality once discretized, distributed nature, and the requirement for low-latency feedback control. Reinforcement Learning (RL), particularly Deep RL (DRL), has recently emerged as a promising control paradigm for such systems, demonstrating exceptional capabilities in managing high-dimensional, nonlinear dynamics. However, DRL faces challenges including sample inefficiency, robustness issues, and an overall lack of interpretability. To address these issues, we propose a data-efficient, interpretable, and scalable Dyna-style Model-Based RL framework for PDE control, combining the Sparse Identification of Nonlinear Dynamics with Control (SINDy-C) algorithm and an autoencoder (AE) framework for the sake of dimensionality reduction of PDE states and actions. This novel approach enables fast rollouts, reducing the need for extensive environment interactions, and provides an interpretable latent space representation of the PDE forward dynamics. We validate our method on two PDE problems describing fluid flows - namely, the 1D Burgers equation and 2D Navier-Stokes equations - comparing it against a model-free baseline, and carrying out an extensive analysis of the learned dynamics.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c107d494b9e41a2fd3b5463f8d71bd772630ca72" target='_blank'>
-              Spectrally Informed Learning of Fluid Flows
+          <td><a href="https://www.semanticscholar.org/paper/c4afec07162a7a857a1ad05397d56c6a677f4b00" target='_blank'>
+              Interpretable and Efficient Data-driven Discovery and Control of Distributed Systems
               </a>
             </td>
           <td>
-            Benjamin D. Shaffer, Jeremy R. Vorenberg, M. A. Hsieh
+            Florian Wolf, Nicolò Botteghi, Urban Fasel, Andrea Manzoni
           </td>
-          <td>2024-08-26</td>
+          <td>2024-11-06</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>2</td>
+          <td>11</td>
         </tr>
 
-        <tr id="This paper presents a sequence of two approaches for the data-driven control-oriented modeling of networked systems, i.e., the systems that involve many interacting dynamical components. First, a novel deep learning approach named the weak Latent Dynamics Model (wLDM) is developed for learning generic nonlinear dynamics with control. Leveraging the weak form, the wLDM enables more numerically stable and computationally efficient training as well as more accurate prediction, when compared to conventional methods such as neural ordinary differential equations. Building upon the wLDM framework, we propose the weak Graph Koopman Bilinear Form (wGKBF) model, which integrates geometric deep learning and Koopman theory to learn latent space dynamics for networked systems, especially for the challenging cases having multiple timescales. The effectiveness of the wLDM framework and wGKBF model are demonstrated on three example systems of increasing complexity - a controlled double pendulum, the stiff Brusselator dynamics, and an electrified aircraft energy system. These numerical examples show that the wLDM and wGKBF achieve superior predictive accuracy and training efficiency as compared to baseline models. Parametric studies provide insights into the effects of hyperparameters in the weak form. The proposed framework shows the capability to efficiently capture control-dependent dynamics in these systems, including stiff dynamics and multi-physics interactions, offering a promising direction for learning control-oriented models of complex networked systems.">
+        <tr id="This paper introduces the Parsimonious Dynamic Mode Decomposition (parsDMD), a novel algorithm designed to automatically select an optimally sparse subset of dynamic modes for both spatiotemporal and purely temporal data. By incorporating time-delay embedding and leveraging Orthogonal Matching Pursuit (OMP), parsDMD ensures robustness against noise and effectively handles complex, nonlinear dynamics. The algorithm is validated on a diverse range of datasets, including standing wave signals, identifying hidden dynamics, fluid dynamics simulations (flow past a cylinder and transonic buffet), and atmospheric sea-surface temperature (SST) data. ParsDMD addresses a significant limitation of the traditional sparsity-promoting DMD (spDMD), which requires manual tuning of sparsity parameters through a rigorous trial-and-error process to balance between single-mode and all-mode solutions. In contrast, parsDMD autonomously determines the optimally sparse subset of modes without user intervention, while maintaining minimal computational complexity. Comparative analyses demonstrate that parsDMD consistently outperforms spDMD by providing more accurate mode identification and effective reconstruction in noisy environments. These advantages render parsDMD an effective tool for real-time diagnostics, forecasting, and reduced-order model construction across various disciplines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c2168dc84b931e82498692c0f9277bed180e91df" target='_blank'>
-              Learning Networked Dynamical System Models with Weak Form and Graph Neural Networks
+          <td><a href="https://www.semanticscholar.org/paper/ababf3f8781fc0d38c9f1a58736153a9dd584068" target='_blank'>
+              Parsimonious Dynamic Mode Decomposition: A Robust and Automated Approach for Optimally Sparse Mode Selection in Complex Systems
               </a>
             </td>
           <td>
-            Yin Yu, Daning Huang, Seho Park, H. Pangborn
+            Arpan Das, Pier Marzocca, O. Levinski
           </td>
-          <td>2024-07-23</td>
+          <td>2024-10-22</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>11</td>
+          <td>9</td>
         </tr>
 
-        <tr id="This paper compared physics-informed neural network (PINN), conventional neural network (NN) and traditional numerical discretization methods on solving differential equations (DEs) through literature investigation and experimental validation. We focused on the soft-constrained PINN approach and formalized its mathematical framework and computational flow for solving Ordinary DEs and Partial DEs (ODEs/PDEs). The working mechanism and its accuracy and efficiency were experimentally verified by solving typical linear and non-linear oscillator ODEs. We demonstrate that the DeepXDE-based implementation of PINN is not only light code and efficient in training, but also flexible across CPU/GPU platforms. PINN greatly reduces the need for labeled data: when the nonlinearity of the ODE is weak, a very small amount of supervised training data plus a few unsupervised collocation points are sufficient to predict the solution; in the minimalist case, only one or two training points (with initial values) are needed for first- or second-order ODEs, respectively. We also find that, with the aid of collocation points and the use of physical information, PINN has the ability to extrapolate data outside the time domain of the training set, and especially is robust to noisy data, thus with enhanced generalization capabilities. Training is accelerated when the gains obtained along with the reduction in the amount of data outweigh the delay caused by the increase in the loss function terms. The soft-constrained PINN can easily impose a physical law (e.g., conservation of energy) constraint by adding a regularization term to the total loss function, thus improving the solution performance to ODEs that obey this physical law. Furthermore, PINN can also be used for stiff ODEs, PDEs, and other types of DEs, and is becoming a favorable catalyst for the era of Digital Twins.">
+        <tr id="Complex systems in physics, chemistry, and biology that evolve over time with inherent randomness are typically described by stochastic differential equations (SDEs). A fundamental challenge in science and engineering is to determine the governing equations of a complex system from snapshot data. Traditional equation discovery methods often rely on stringent assumptions, such as the availability of the trajectory information or time-series data, and the presumption that the underlying system is deterministic. In this work, we introduce a data-driven, simulation-free framework, called Sparse Identification of Differential Equations from Snapshots (SpIDES), that discovers the governing equations of a complex system from snapshots by utilizing the advanced machine learning techniques to perform three essential steps: probability flow reconstruction, probability density estimation, and Bayesian sparse identification. We validate the effectiveness and robustness of SpIDES by successfully identifying the governing equation of an over-damped Langevin system confined within two potential wells. By extracting interpretable drift and diffusion terms from the SDEs, our framework provides deeper insights into system dynamics, enhances predictive accuracy, and facilitates more effective strategies for managing and simulating stochastic systems.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/7e1491f0dafa40309998996d32fcbb07f21c3127" target='_blank'>
-              Solving Oscillator Ordinary Differential Equations via Soft-constrained Physics-informed Neural Network with Small Data
+          <td><a href="https://www.semanticscholar.org/paper/4406159646424ab84642702029afaf833d2a7a82" target='_blank'>
+              Governing equation discovery of a complex system from snapshots
               </a>
             </td>
           <td>
-            Kai-liang Lu, Yu-meng Su, Zhuo Bi, Cheng Qiu, Wen-jun Zhang
+            Qunxi Zhu, Bolin Zhao, Jingdong Zhang, Peiyang Li, Wei Lin
           </td>
-          <td>2024-08-19</td>
+          <td>2024-10-22</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>10</td>
         </tr>
 
-        <tr id="Time-varying linear state-space models are powerful tools for obtaining mathematically interpretable representations of neural signals. For example, switching and decomposed models describe complex systems using latent variables that evolve according to simple locally linear dynamics. However, existing methods for latent variable estimation are not robust to dynamical noise and system nonlinearity due to noise-sensitive inference procedures and limited model formulations. This can lead to inconsistent results on signals with similar dynamics, limiting the model's ability to provide scientific insight. In this work, we address these limitations and propose a probabilistic approach to latent variable estimation in decomposed models that improves robustness against dynamical noise. Additionally, we introduce an extended latent dynamics model to improve robustness against system nonlinearities. We evaluate our approach on several synthetic dynamical systems, including an empirically-derived brain-computer interface experiment, and demonstrate more accurate latent variable inference in nonlinear systems with diverse noise conditions. Furthermore, we apply our method to a real-world clinical neurophysiology dataset, illustrating the ability to identify interpretable and coherent structure where previous models cannot.">
+        <tr id="Dynamical systems theory has long provided a foundation for understanding evolving phenomena across scientific domains. Yet, the application of this theory to complex real-world systems remains challenging due to issues in mathematical modeling, nonlinearity, and high dimensionality. In this work, we introduce a data-driven computational framework to derive low-dimensional linear models for nonlinear dynamical systems directly from raw experimental data. This framework enables global stability analysis through interpretable linear models that capture the underlying system structure. Our approach employs time-delay embedding, physics-informed deep autoencoders, and annealing-based regularization to identify novel low-dimensional coordinate representations, unlocking insights across a variety of simulated and previously unstudied experimental dynamical systems. These new coordinate representations enable accurate long-horizon predictions and automatic identification of intricate invariant sets while providing empirical stability guarantees. Our method offers a promising pathway to analyze complex dynamical behaviors across fields such as physics, climate science, and engineering, with broad implications for understanding nonlinear systems in the real world.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/a880c22bf304a23aeae9d7e165c0f42a99f4dd1b" target='_blank'>
-              Probabilistic Decomposed Linear Dynamical Systems for Robust Discovery of Latent Neural Dynamics
+          <td><a href="https://www.semanticscholar.org/paper/ac835425463bfb6a585132a06ba8fbce9643ce8d" target='_blank'>
+              Automated Global Analysis of Experimental Dynamics through Low-Dimensional Linear Embeddings
               </a>
             </td>
           <td>
-            Yenho Chen, Noga Mudrik, Kyle A. Johnsen, Sankaraleengam (Sankar) Alagapan, Adam S. Charles, Christopher J. Rozell
+            Samuel A. Moore, Brian P. Mann, Boyuan Chen
           </td>
-          <td>2024-08-29</td>
+          <td>2024-11-01</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>19</td>
+          <td>2</td>
         </tr>
 
-        <tr id="Machine learning techniques have recently been of great interest for solving differential equations. Training these models is classically a data-fitting task, but knowledge of the expression of the differential equation can be used to supplement the training objective, leading to the development of physics-informed scientific machine learning. In this article, we focus on one class of models called nonlinear vector autoregression (NVAR) to solve ordinary differential equations (ODEs). Motivated by connections to numerical integration and physics-informed neural networks, we explicitly derive the physics-informed NVAR (piNVAR) which enforces the right-hand side of the underlying differential equation regardless of NVAR construction. Because NVAR and piNVAR completely share their learned parameters, we propose an augmented procedure to jointly train the two models. Then, using both data-driven and ODE-driven metrics, we evaluate the ability of the piNVAR model to predict solutions to various ODE systems, such as the undamped spring, a Lotka-Volterra predator-prey nonlinear model, and the chaotic Lorenz system.">
+        <tr id="Understanding the governing rules of complex biological systems remains a significant challenge due to the nonlinear, high-dimensional nature of biological data. In this study, we present CLERA, a novel end-to-end computational framework designed to uncover parsimonious dynamical models and identify active gene programs from single-cell RNA sequencing data. By integrating a supervised autoencoder architecture with Sparse Identification of Nonlinear Dynamics, CLERA leverages prior knowledge to simultaneously extract related low-dimensional embeddings and uncovers the underlying dynamical systems that drive the processes. Through the analysis of both synthetic and biological datasets, CLERA demonstrates robust performance in reconstructing gene expression dynamics, identifying key regulatory genes, and capturing temporal patterns across distinct cell types. CLERA’s ability to generate dynamic interaction networks, combined with network rewiring using Personalized PageRank to highlight central genes and active gene programs, offers new insights into the complex regulatory mechanisms underlying cellular processes.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/59096640a40ed6b70acbbfdaca5a505cd1330bb4" target='_blank'>
-              Physics-informed nonlinear vector autoregressive models for the prediction of dynamical systems
+          <td><a href="https://www.semanticscholar.org/paper/ee363ce2688ff6b072bec9a96de6ee85cb7b46b3" target='_blank'>
+              Discovering Governing Equations of Biological Systems through Representation Learning and Sparse Model Discovery
               </a>
             </td>
           <td>
-            James H. Adler, Samuel Hocking, Xiaozhe Hu, Shafiqul Islam
+            Mehrshad Sadria, Vasu Swaroop
           </td>
-          <td>2024-07-25</td>
-          <td>ArXiv</td>
+          <td>2024-09-23</td>
+          <td>bioRxiv</td>
           <td>0</td>
-          <td>2</td>
+          <td>8</td>
         </tr>
 
-        <tr id="We propose a novel learning framework for Koopman operator of nonlinear dynamical systems that is informed by the governing equation and guarantees long-time stability and robustness to noise. In contrast to existing frameworks where either ad-hoc observables or blackbox neural networks are used to construct observables in the extended dynamic mode decomposition (EDMD), our observables are informed by governing equations via Polyflow. To improve the noise robustness and guarantee long-term stability, we designed a stable parameterization of the Koopman operator together with a progressive learning strategy for roll-out recurrent loss. To further improve model performance in the phase space, a simple iterative strategy of data augmentation was developed. Numerical experiments of prediction and control of classic nonlinear systems with ablation study showed the effectiveness of the proposed techniques over several state-of-the-art practices.">
+        <tr id="Complex chaotic dynamics, seen in natural and industrial systems like turbulent flows and weather patterns, often span vast spatial domains with interactions across scales. Accurately capturing these features requires a high-dimensional state space to resolve all the time and spatial scales. For dissipative systems the dynamics lie on a finite-dimensional manifold with fewer degrees of freedom. Thus, by building reduced-order data-driven models in manifold coordinates, we can capture the essential behavior of chaotic systems. Unfortunately, these tend to be formulated globally rendering them less effective for large spatial systems. In this context, we present a data-driven low-dimensional modeling approach to tackle the complexities of chaotic motion, Markovian dynamics, multi-scale behavior, and high numbers of degrees of freedom within large spatial domains. Our methodology involves a parallel scheme of decomposing a spatially extended system into a sequence of local `patches', and constructing a set of coupled, local low-dimensional dynamical models for each patch. Here, we choose to construct the set of local models using autoencoders (for constructing the low-dimensional representation) and neural ordinary differential equations, NODE, for learning the evolution equation. Each patch, or local model, shares the same underlying functions (e.g., autoencoders and NODEs) due to the spatial homogeneity of the underlying systems we consider. We apply this method to the Kuramoto-Sivashinsky equation and 2D Kolmogorov flow, and reduce state dimension by up to two orders of magnitude while accurately capturing both short-term dynamics and long-term statistics.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/33d038a0e7775662768898c432284e433922c891" target='_blank'>
-              Learning Noise-Robust Stable Koopman Operator for Control with Physics-Informed Observables
+          <td><a href="https://www.semanticscholar.org/paper/f1d9b7cb4acfb333508b09fbb7b41c8e84bdcb2a" target='_blank'>
+              Data-driven prediction of large-scale spatiotemporal chaos with distributed low-dimensional models
               </a>
             </td>
           <td>
-            Shahriar Akbar Sakib, Shaowu Pan
+            Ricardo Constante-Amores, Alec J. Linot, Michael D. Graham
           </td>
-          <td>2024-08-13</td>
+          <td>2024-10-02</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>8</td>
         </tr>
 
-        <tr id="Stochastic collocation (SC) is a well‐known non‐intrusive method of constructing surrogate models for uncertainty quantification. In dynamical systems, SC is especially suited for full‐field uncertainty propagation that characterizes the distributions of the high‐dimensional solution fields of a model with stochastic input parameters. However, due to the highly nonlinear nature of the parameter‐to‐solution map in even the simplest dynamical systems, the constructed SC surrogates are often inaccurate. This work presents an alternative approach, where we apply the SC approximation over the dynamics of the model, rather than the solution. By combining the data‐driven sparse identification of nonlinear dynamics framework with SC, we construct dynamics surrogates and integrate them through time to construct the surrogate solutions. We demonstrate that the SC‐over‐dynamics framework leads to smaller errors, both in terms of the approximated system trajectories as well as the model state distributions, when compared against full‐field SC applied to the solutions directly. We present numerical evidence of this improvement using three test problems: a chaotic ordinary differential equation, and two partial differential equations from solid mechanics.">
+        <tr id="We propose the use of the Extended Kalman Filter (EKF) for online data assimilation and update of a dynamic model, preliminary identified through the Sparse Identification of Nonlinear Dynamics (SINDy). This data-driven technique may avoid biases due to incorrect modelling assumptions and exploits SINDy to approximate the system dynamics leveraging a predefined library of functions, where active terms are selected and weighted by a sparse set of coefficients. This results in a physically-sound and interpretable dynamic model allowing to reduce epistemic uncertainty often affecting machine learning approaches. Treating the SINDy model coefficients as random variables, we propose to update them while acquiring (possibly noisy) system measurements, thus enabling the online identification of time-varying systems. These changes can stem from, e.g., varying operational conditions or unforeseen events. The EKF performs model adaptation through joint state-parameters estimation, with the Jacobian matrices required to computed the model sensitivity inexpensively evaluated from the SINDy model formulation. The effectiveness of this approach is demonstrated through three case studies: (i) a Lokta-Volterra model in which all parameters simultaneously evolve during the observation period; (ii) a Selkov model where the system undergoes a bifurcation not seen during the SINDy training; (iii) a MEMS arch exhibiting a 1:2 internal resonance. The ability of EKF of recovering inactivated functional terms from the SINDy library, or discarding unnecessary contribution, is also highlighted. Based on the presented applications, this method shows strong promise for handling time-varying nonlinear dynamic systems possibly experiencing bifurcating behaviours.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/dc0b58e254bb512300814c52b4a101046377d6e4" target='_blank'>
-              Accurate data‐driven surrogates of dynamical systems for forward propagation of uncertainty
+          <td><a href="https://www.semanticscholar.org/paper/2d108c654cd0caef592cad2586543c394a27e334" target='_blank'>
+              Online learning in bifurcating dynamic systems via SINDy and Kalman filtering
               </a>
             </td>
           <td>
-            Saibal De, Reese E. Jones, H. Kolla
+            Luca Rosafalco, Paolo Conti, Andrea Manzoni, Stefano Mariani, A. Frangi
           </td>
-          <td>2024-08-03</td>
-          <td>International Journal for Numerical Methods in Engineering</td>
+          <td>2024-11-07</td>
+          <td>ArXiv</td>
           <td>0</td>
-          <td>30</td>
+          <td>8</td>
         </tr>
 
-        <tr id="Data-driven discovery of partial differential equations (PDEs) has emerged as a promising approach for deriving governing physics when domain knowledge about observed data is limited. Despite recent progress, the identification of governing equations and their parametric dependencies using conventional information criteria remains challenging in noisy situations, as the criteria tend to select overly complex PDEs. In this paper, we introduce an extension of the uncertainty-penalized Bayesian information criterion (UBIC), which is adapted to solve parametric PDE discovery problems efficiently without requiring computationally expensive PDE simulations. This extended UBIC uses quantified PDE uncertainty over different temporal or spatial points to prevent overfitting in model selection. The UBIC is computed with data transformation based on power spectral densities to discover the governing parametric PDE that truly captures qualitative features in frequency space with a few significant terms and their parametric dependencies (i.e., the varying PDE coefficients), evaluated with confidence intervals. Numerical experiments on canonical PDEs demonstrate that our extended UBIC can identify the true number of terms and their varying coefficients accurately, even in the presence of noise. The code is available at \url{https://github.com/Pongpisit-Thanasutives/parametric-discovery}.">
+        <tr id="Data-driven model discovery of complex dynamical systems is typically done using sparse optimization, but it has a fundamental limitation: sparsity in that the underlying governing equations of the system contain only a small number of elementary mathematical terms. Examples where sparse optimization fails abound, such as the classic Ikeda or optical-cavity map in nonlinear dynamics and a large variety of ecosystems. Exploiting the recently articulated Kolmogorov-Arnold networks, we develop a general model-discovery framework for any dynamical systems including those that do not satisfy the sparsity condition. In particular, we demonstrate non-uniqueness in that a large number of approximate models of the system can be found which generate the same invariant set with the correct statistics such as the Lyapunov exponents and Kullback-Leibler divergence. An analogy to shadowing of numerical trajectories in chaotic systems is pointed out.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/d605721a21ed5e6e69cc14a1ab3eb9a559099ea9" target='_blank'>
-              Adaptation of uncertainty-penalized Bayesian information criterion for parametric partial differential equation discovery
+          <td><a href="https://www.semanticscholar.org/paper/cf52e922a7b95ad02966f5ab74138023b406c038" target='_blank'>
+              Data-driven model discovery with Kolmogorov-Arnold networks
               </a>
             </td>
           <td>
-            Pongpisit Thanasutives, Ken-ichi Fukui
+            Mohammadamin Moradi, Shirin Panahi, E. Bollt, Ying-Cheng Lai
           </td>
-          <td>2024-08-15</td>
+          <td>2024-09-23</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>3</td>
+          <td>2</td>
+          <td>37</td>
         </tr>
 
-        <tr id="Structural dynamics models with nonlinear stiffness appear, for example, when analyzing systems with nonlinear material behavior or undergoing large deformations. For complex systems, these models become too large for real-time applications or multi-query workflows. Hence, model reduction is needed. However, the mathematical operators of these models are often not available since, as is common in industry practice, the models are constructed using commercial simulation software. In this work, we propose an operator inference-based approach aimed at inferring, from data generated by the simulation model, reduced-order models (ROMs) of structural dynamics systems with stiffness terms represented by polynomials of arbitrary degree. To ensure physically meaningful models, we impose constraints on the inference such that the model is guaranteed to exhibit stability properties. Convexity of the optimization problem associated with the inference is maintained by applying a sum-of-squares relaxation to the polynomial term. To further reduce the size of the ROM and improve numerical conditioning of the inference, we also propose a novel clustering-based sparsification of the polynomial term. We validate the proposed method on several numerical examples, including a representative 3D Finite Element Model (FEM) of a steel piston rod.">
+        <tr id="Discovering explicit governing equations of stochastic dynamical systems with both (Gaussian) Brownian noise and (non-Gaussian) L\'evy noise from data is chanllenging due to possible intricate functional forms and the inherent complexity of L\'evy motion. This present research endeavors to develop an evolutionary symbol sparse regression (ESSR) approach to extract non-Gaussian stochastic dynamical systems from sample path data, based on nonlocal Kramers-Moyal formulas, genetic programming, and sparse regression. More specifically, the genetic programming is employed to generate a diverse array of candidate functions, the sparse regression technique aims at learning the coefficients associated with these candidates, and the nonlocal Kramers-Moyal formulas serve as the foundation for constructing the fitness measure in genetic programming and the loss function in sparse regression. The efficacy and capabilities of this approach are showcased through its application to several illustrative models. This approach stands out as a potent instrument for deciphering non-Gaussian stochastic dynamics from available datasets, indicating a wide range of applications across different fields.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/12474e97123f0f45095cbb219da196a32731873d" target='_blank'>
-              Stable Sparse Operator Inference for Nonlinear Structural Dynamics
+          <td><a href="https://www.semanticscholar.org/paper/92a8fe5ae6040e98e4e335d324524edb1ada08a5" target='_blank'>
+              An evolutionary approach for discovering non-Gaussian stochastic dynamical systems based on nonlocal Kramers-Moyal formulas
               </a>
             </td>
           <td>
-            P. D. Boef, Diana Manvelyan, Jos Maubach, W. Schilders, N. Wouw
+            Yang Li, Shengyuan Xu, Jinqiao Duan
           </td>
-          <td>2024-07-31</td>
+          <td>2024-09-29</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>47</td>
+          <td>0</td>
         </tr>
 
-        <tr id="Koopman operator theory offers a rigorous treatment of dynamics and has been emerging as a powerful modeling and learning-based control method enabling significant advancements across various domains of robotics. Due to its ability to represent nonlinear dynamics as a linear operator, Koopman theory offers a fresh lens through which to understand and tackle the modeling and control of complex robotic systems. Moreover, it enables incremental updates and is computationally inexpensive making it particularly appealing for real-time applications and online active learning. This review comprehensively presents recent research results on advancing Koopman operator theory across diverse domains of robotics, encompassing aerial, legged, wheeled, underwater, soft, and manipulator robotics. Furthermore, it offers practical tutorials to help new users get started as well as a treatise of more advanced topics leading to an outlook on future directions and open research questions. Taken together, these provide insights into the potential evolution of Koopman theory as applied to the field of robotics.">
+        <tr id="Motivated by the surge of interest in Koopman operator theory, we propose a machine-learning alternative based on a functional Bayesian perspective for operator-theoretic modeling of unknown, data-driven, nonlinear dynamical systems. This formulation is directly done in an infinite-dimensional space of linear operators or Hilbert space with universal approximation property. The theory of reproducing kernel Hilbert space (RKHS) allows the lifting of nonlinear dynamics to a potentially infinite-dimensional space via linear embeddings, where a general nonlinear function is represented as a set of linear functions or operators in the functional space. This allows us to apply classical linear Bayesian methods such as the Kalman filter directly in the Hilbert space, yielding nonlinear solutions in the original input space. This kernel perspective on the Koopman operator offers two compelling advantages. First, the Hilbert space can be constructed deterministically, agnostic to the nonlinear dynamics. The Gaussian kernel is universal, approximating uniformly an arbitrary continuous target function over any compact domain. Second, Bayesian filter is an adaptive, linear minimum-variance algorithm, allowing the system to update the Koopman operator and continuously track the changes across an extended period of time, ideally suited for modern data-driven applications such as real-time machine learning using streaming data. In this paper, we present several practical implementations to obtain a finite-dimensional approximation of the functional Bayesian filter (FBF). Due to the rapid decay of the Gaussian kernel, excellent approximation is obtained with a small dimension. We demonstrate that this practical approach can obtain accurate results and outperform finite-dimensional Koopman decomposition.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/14f58618f049522401ec48897f882dcc38849d2b" target='_blank'>
-              Koopman Operators in Robot Learning
+          <td><a href="https://www.semanticscholar.org/paper/c6f90946466ada48fed9585f5ab005d325d95a21" target='_blank'>
+              Kernel Operator-Theoretic Bayesian Filter for Nonlinear Dynamical Systems
               </a>
             </td>
           <td>
-            Lu Shi, Masih Haseli, Giorgos Mamakoukas, Daniel Bruder, Ian Abraham, Todd Murphey, Jorge Cortes, Konstantinos Karydis
+            Kan Li, Jos'e C. Pr'incipe
           </td>
-          <td>2024-08-08</td>
+          <td>2024-10-31</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>9</td>
+          <td>2</td>
         </tr>
 
-        <tr id="This paper presents a probabilistic approach to represent and quantify model-form uncertainties in the reduced-order modeling of complex systems using operator inference techniques. Such uncertainties can arise in the selection of an appropriate state-space representation, in the projection step that underlies many reduced-order modeling methods, or as a byproduct of considerations made during training, to name a few. Following previous works in the literature, the proposed method captures these uncertainties by expanding the approximation space through the randomization of the projection matrix. This is achieved by combining Riemannian projection and retraction operators - acting on a subset of the Stiefel manifold - with an information-theoretic formulation. The efficacy of the approach is assessed on canonical problems in fluid mechanics by identifying and quantifying the impact of model-form uncertainties on the inferred operators.">
+        <tr id="Deep learning is widely used to predict complex dynamical systems in many scientific and engineering areas. However, the black-box nature of these deep learning models presents significant challenges for carrying out simultaneous data assimilation (DA), which is a crucial technique for state estimation, model identification, and reconstructing missing data. Integrating ensemble-based DA methods with nonlinear deep learning models is computationally expensive and may suffer from large sampling errors. To address these challenges, we introduce a deep learning framework designed to simultaneously provide accurate forecasts and efficient DA. It is named Conditional Gaussian Koopman Network (CGKN), which transforms general nonlinear systems into nonlinear neural differential equations with conditional Gaussian structures. CGKN aims to retain essential nonlinear components while applying systematic and minimal simplifications to facilitate the development of analytic formulae for nonlinear DA. This allows for seamless integration of DA performance into the deep learning training process, eliminating the need for empirical tuning as required in ensemble methods. CGKN compensates for structural simplifications by lifting the dimension of the system, which is motivated by Koopman theory. Nevertheless, CGKN exploits special nonlinear dynamics within the lifted space. This enables the model to capture extreme events and strong non-Gaussian features in joint and marginal distributions with appropriate uncertainty quantification. We demonstrate the effectiveness of CGKN for both prediction and DA on three strongly nonlinear and non-Gaussian turbulent systems: the projected stochastic Burgers--Sivashinsky equation, the Lorenz 96 system, and the El Ni\~no-Southern Oscillation. The results justify the robustness and computational efficiency of CGKN.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/9d2c7ce971b6feea68b5b1b339389e2f4d4a0e62" target='_blank'>
-              Learning Latent Space Dynamics with Model-Form Uncertainties: A Stochastic Reduced-Order Modeling Approach
+          <td><a href="https://www.semanticscholar.org/paper/394d2bab1a7b169a1a96f765eea12d6aad03bac4" target='_blank'>
+              CGKN: A Deep Learning Framework for Modeling Complex Dynamical Systems and Efficient Data Assimilation
               </a>
             </td>
           <td>
-            Jin Yi Yong, Rudy Geelen, Johann Guilleminot
+            Chuanqi Chen, Nan Chen, Yinling Zhang, Jin-Long Wu
           </td>
-          <td>2024-08-30</td>
+          <td>2024-10-26</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>3</td>
         </tr>
 
-        <tr id="This article studies two particular algorithms, a Relaxation Least Squares (RLS) algorithm and a Relaxation Newton Iteration (RNI) scheme , for reconstructing unknown parameters in dissipative dynamical systems. Both algorithms are based on a continuous data assimilation (CDA) algorithm for state reconstruction of A. Azouani, E. Olson, and E.S. Titi \cite{Azouani_Olson_Titi_2014}. Due to the CDA origins of these parameter recovery algorithms, these schemes provide on-the-fly reconstruction, that is, as data is collected, of unknown state and parameters simultaneously. It is shown how both algorithms give way to a robust general framework for simultaneous state and parameter estimation. In particular, we develop a general theory, applicable to a large class of dissipative dynamical systems, which identifies structural and algorithmic conditions under which the proposed algorithms achieve reconstruction of the true parameters. The algorithms are implemented on a high-dimensional two-layer Lorenz 96 model, where the theoretical conditions of the general framework are explicitly verifiable. They are also implemented on the two-dimensional Rayleigh-B\'enard convection system to demonstrate the applicability of the algorithms beyond the finite-dimensional setting. In each case, systematic numerical experiments are carried out probing the efficacy of the proposed algorithms, in addition to the apparent benefits and drawbacks between them.">
+        <tr id="In scientific machine learning, the task of identifying partial differential equations accurately from sparse and noisy data poses a significant challenge. Current sparse regression methods may identify inaccurate equations on sparse and noisy datasets and are not suitable for varying coefficients. To address this issue, we propose a hybrid framework that combines two alternating direction optimization phases: discovery and embedding. The discovery phase employs current well-developed sparse regression techniques to preliminarily identify governing equations from observations. The embedding phase implements a recurrent convolutional neural network (RCNN), enabling efficient processes for time-space iterations involved in discretized forms of wave equation. The RCNN model further optimizes the imperfect sparse regression results to obtain more accurate functional terms and coefficients. Through alternating update of discovery-embedding phases, essential physical equations can be robustly identified from noisy and low-resolution measurements. To assess the performance of proposed framework, numerical experiments are conducted on various scenarios involving wave equation in elastic/viscoelastic and homogeneous/inhomogeneous media. The results demonstrate that the proposed method exhibits excellent robustness and accuracy, even when faced with high levels of noise and limited data availability in both spatial and temporal domains.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/77c455250b844b0b16ee5f82ee489059f0376d4f" target='_blank'>
-              Relaxation-based schemes for on-the-fly parameter estimation in dissipative dynamical systems
+          <td><a href="https://www.semanticscholar.org/paper/d81366e7c59f72832148f69559b35cc6cad72c43" target='_blank'>
+              Discovery and inversion of the viscoelastic wave equation in inhomogeneous media
               </a>
             </td>
           <td>
-            Vincent R. Martinez, Jacob Murri, J. Whitehead
+            Su Chen, Yi Ding, Hiroe Miyake, Xiaojun Li
           </td>
-          <td>2024-08-26</td>
+          <td>2024-09-27</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>2</td>
         </tr>
 
-        <tr id="While linear systems are well-understood, no explicit solution exists for general nonlinear systems. Thus, it is desirable to make the understanding of linear system available in the nonlinear setting. This motivates the search for linearization techniques which are able to represent the nonlinear system by an equivalent, linear system. While much progress has been made extending various linearization techniques to larger domains and more delicate attractor geometries, the limitations of such techniques for truly nonlinear dynamics, such as dynamics with coexisting attractors, have been pointed out recently. In this work, we show that genuinely nonlinear dynamics can be globally linearized. To this end, we investigate systems with a continuous spectrum, a limit cycle, and coexisting solutions and explicitly construct linear systems mimicking these nonlinear behaviors as close as possible. We approximate the transformations between linear and nonlinear system with deep neural networks. The obtained linearizations are finite dimensional exceeding the phase space dimension of the underlying linear system by one at most.">
+        <tr id="Ultrafast optics is driven by a myriad of complex nonlinear dynamics. The ubiquitous presence of governing equations in the form of partial integro-differential equations (PIDE) necessitates the need for advanced computational tools to understand the underlying physical mechanisms. From the experimental perspective, signal-to-noise ratio and availability of measurable data, accounts for a bottle neck in numerical and data-driven modeling methods. In this paper we extend the application of the physics informed neural network (PINN) architecture to include prior knowledge in both the physical and Fourier domain. We demonstrate our Fourier Domain PINN (FD-PINN) in two distinct forms. The Continuous time FD-PINN is used to predict accurate solutions to the Generalized Pulse Propagation Equation, which includes the complete delayed nonlinear response, in the data-starved and noisy regime. We extend the architecture to the Discrete time FD-PINN to recover the delayed-response physics from spatially separated measurement points. We conducted the first systematic study of the effect of SNR on the spatiotemporal field prediction as well as physics discovery. Our architecture ensures high fidelity predictive modeling and hidden physics recovery for applications such as image reconstruction, pulse characterization and shaping, as well as hidden parameter discovery. The benefits of the FD-PINN for ultrafast nonlinear optics make it immediately experimentally deployable. FD-PINN represents the next generation of tools to study optical phenomena both through modeling and measurements for both forward and inverse problems.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2196ed95f5d349e07c7af94fd99240eec34dfdb2" target='_blank'>
-              Learning Global Linear Representations of Truly Nonlinear Dynamics
+          <td><a href="https://www.semanticscholar.org/paper/83e7a6c9c52d8790d3533dd032d3b73d1802c725" target='_blank'>
+              Fourier Domain Physics Informed Neural Network
               </a>
             </td>
           <td>
-            Thomas Breunung, F. Kogelbauer
+            Jonathan Musgrave, Shu-Wei Huang
           </td>
-          <td>2024-08-06</td>
+          <td>2024-09-30</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>6</td>
+          <td>1</td>
         </tr>
 
-        <tr id="Physics-Informed Neural Networks (PINNs) have revolutionized solving differential equations by integrating physical laws into neural network training. This paper explores PINNs for open-loop optimal control problems (OCPs) with incomplete information, such as unknown initial conditions and sparse boundary data. We derive optimality conditions from the Lagrangian multipliers and use a neural network to predict the state, adjoint, and control variables. In contrast with previous methods, our approach integrates these elements into a single neural network and addresses scenarios with consistently limited data. Specifically, we address the study of partially unknown equations identifying underlying parameters online by searching for the optimal solution. Numerical examples show the effectiveness of the proposed method even in scenarios characterized by a considerable lack of information.">
+        <tr id="None">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/1de965a4fc10f47113e0b8b25e5ae6627f2c9cbb" target='_blank'>
-              A PINN approach for the online identification and control of unknown PDEs
+          <td><a href="https://www.semanticscholar.org/paper/037d5b01fa6bf3a903830fd8399f38895b65e5c8" target='_blank'>
+              Reinforcement learning-based estimation for spatio-temporal systems
               </a>
             </td>
           <td>
-            Alessandro Alla, Giulia Bertaglia, Elisa Calzola
+            Saviz Mowlavi, M. Benosman
           </td>
-          <td>2024-08-06</td>
-          <td>ArXiv</td>
+          <td>2024-09-28</td>
+          <td>Scientific Reports</td>
           <td>0</td>
-          <td>1</td>
+          <td>23</td>
         </tr>
 
-        <tr id="We present a pragmatic approach to the sparse identification of nonlinear dynamics for systems with discrete delays. It relies on approximating the underlying delay model with a system of ordinary differential equations via pseudospectral collocation. To minimize the reconstruction error, the new strategy avoids optimizing all possible multiple unknown delays, identifying only the maximum one. The computational burden is thus greatly reduced, improving the performance of recent implementations that work directly on the delay system.">
+        <tr id="Dynamical systems form the foundation of scientific discovery, traditionally modeled with predefined state variables such as the angle and angular velocity, and differential equations such as the equation of motion for a single pendulum. We propose an approach to discover a set of state variables that preserve the smoothness of the system dynamics and to construct a vector field representing the system's dynamics equation, automatically from video streams without prior physical knowledge. The prominence and effectiveness of the proposed approach are demonstrated through both quantitative and qualitative analyses of various dynamical systems, including the prediction of characteristic frequencies and the identification of chaotic and limit cycle behaviors. This shows the potential of our approach to assist human scientists in scientific discovery.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/af6a7301b4680a556da5635cdd6aedbb64f047b0" target='_blank'>
-              Sparse identification of time delay systems via pseudospectral collocation
+          <td><a href="https://www.semanticscholar.org/paper/6e5457c98e86e80dc7ee5217c3b0e18000ab5348" target='_blank'>
+              Automated Discovery of Continuous Dynamics from Videos
               </a>
             </td>
           <td>
-            Enrico Bozzo, Dimitri Breda, Muhammad Tanveer
+            Kuang Huang, Dong Heon Cho, Boyuan Chen
           </td>
-          <td>2024-08-04</td>
+          <td>2024-10-14</td>
           <td>ArXiv</td>
-          <td>0</td>
+          <td>1</td>
           <td>0</td>
         </tr>
 
-        <tr id="Conventional physics-based modeling techniques involve high effort, e.g., time and expert knowledge, while data-driven methods often lack interpretability, structure, and sometimes reliability. To mitigate this, we present a data-driven system identification framework that derives models in the port-Hamiltonian (pH) formulation. This formulation is suitable for multi-physical systems while guaranteeing the useful system theoretical properties of passivity and stability. Our framework combines linear and nonlinear reduction with structured, physics-motivated system identification. In this process, high-dimensional state data obtained from possibly nonlinear systems serves as input for an autoencoder, which then performs two tasks: (i) nonlinearly transforming and (ii) reducing this data onto a low-dimensional latent space. In this space, a linear pH system, that satisfies the pH properties per construction, is parameterized by the weights of a neural network. The mathematical requirements are met by defining the pH matrices through Cholesky factorizations. The neural networks that define the coordinate transformation and the pH system are identified in a joint optimization process to match the dynamics observed in the data while defining a linear pH system in the latent space. The learned, low-dimensional pH system can describe even nonlinear systems and is rapidly computable due to its small size. The method is exemplified by a parametric mass-spring-damper and a nonlinear pendulum example, as well as the high-dimensional model of a disc brake with linear thermoelastic behavior.">
+        <tr id="Partial differential equations, and their chaotic solutions, are pervasive in the modelling of complex systems in engineering, science, and beyond. Data-driven methods can find solutions to partial differential equations with a divide-and-conquer strategy: The solution is sought in a latent space, on which the temporal dynamics are inferred (``latent-space'' approach). This is achieved by, first, compressing the data with an autoencoder, and, second, inferring the temporal dynamics with recurrent neural networks. The overarching goal of this paper is to show that a latent-space approach can not only infer the solution of a chaotic partial differential equation, but it can also predict the stability properties of the physical system. First, we employ the convolutional autoencoder echo state network (CAE-ESN) on the chaotic Kuramoto-Sivashinsky equation for various chaotic regimes. We show that the CAE-ESN (i) finds a low-dimensional latent-space representation of the observations and (ii) accurately infers the Lyapunov exponents and covariant Lyapunov vectors (CLVs) in this low-dimensional manifold for different attractors. Second, we extend the CAE-ESN to a turbulent flow, comparing the Lyapunov spectrum to estimates obtained from Jacobian-free methods. A latent-space approach based on the CAE-ESN effectively produces a latent space that preserves the key properties of the chaotic system, such as Lyapunov exponents and CLVs, thus retaining the geometric structure of the attractor. The latent-space approach based on the CAE-ESN is a reduced-order model that accurately predicts the dynamics of the chaotic system, or, alternatively, it can be used to infer stability properties of chaotic systems from data.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/8c88f9ec4b5bd522ac8e312f90d2801489785951" target='_blank'>
-              Data-driven identification of latent port-Hamiltonian systems
+          <td><a href="https://www.semanticscholar.org/paper/40a24ed82c56c25b5f7fc0acaad378e756d9644b" target='_blank'>
+              Stability analysis of chaotic systems in latent spaces
               </a>
             </td>
           <td>
-            J. Rettberg, Jonas Kneifl, Julius Herb, Patrick Buchfink, J. Fehr, B. Haasdonk
+            Elise Ozalp, L. Magri
           </td>
-          <td>2024-08-15</td>
+          <td>2024-10-01</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>32</td>
+          <td>2</td>
         </tr>
 
-        <tr id="Closure models are widely used in simulating complex multiscale dynamical systems such as turbulence and the earth system, for which direct numerical simulation that resolves all scales is often too expensive. For those systems without a clear scale separation, deterministic and local closure models often lack enough generalization capability, which limits their performance in many real-world applications. In this work, we propose a data-driven modeling framework for constructing stochastic and non-local closure models via conditional diffusion model and neural operator. Specifically, the Fourier neural operator is incorporated into a score-based diffusion model, which serves as a data-driven stochastic closure model for complex dynamical systems governed by partial differential equations (PDEs). We also demonstrate how accelerated sampling methods can improve the efficiency of the data-driven stochastic closure model. The results show that the proposed methodology provides a systematic approach via generative machine learning techniques to construct data-driven stochastic closure models for multiscale dynamical systems with continuous spatiotemporal fields.">
+        <tr id="Discovering mechanisms underlying the behaviors of complex, high dimensional, and nonlinear dynamical systems is a central goal of the natural and synthetic sciences. Breakthroughs in machine learning in concert with increasing capacities for computation and data collection have enabled the use of trajectory measurements for learning predictive models. However, rigorous approaches for interpreting mechanisms from such models remain elusive, and asymptotic prediction accuracy suffers if the model does not capture important state space structures (e.g., attracting invariant sets). These limitations are especially pressing for system-level behaviors such as whole-body locomotion, where discontinuous, transient, and multiscale phenomena are common and prior models are rare. To take the next step towards a theory and practice for dynamical inference of complex multiscale systems in biology and beyond, we introduce VERT, a framework for learning the attracting sets that characterize global system behavior without recourse to learning a global model. Our approach is based on an infinitesimal-local-global (ILG) framework for estimating the proximity of any sampled state to the attracting set, if one exists, with formal accuracy guarantees. We demonstrate our approach on phenomenological and physical oscillators with hierarchical and impulsive dynamics, finding sensitivity to both global and intermediate attractors composed in sequence and parallel. Application of VERT to human running kinematics data reveals insight into control modules that stabilize task-level dynamics, supporting a longstanding neuromechanical control hypothesis. The VERT framework thus enables rigorous inference of underlying dynamical structure even for systems where learning a global dynamics model is impractical or impossible.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/5b117384f237e71d202a61374a42d92e97b0c697" target='_blank'>
-              Data-Driven Stochastic Closure Modeling via Conditional Diffusion Model and Neural Operator
+          <td><a href="https://www.semanticscholar.org/paper/e25805462e292c7645cc99038b9a3d068fa86e7d" target='_blank'>
+              Global dynamical structures from infinitesimal data
               </a>
             </td>
           <td>
-            Xinghao Dong, Chuanqi Chen, Jin-Long Wu
+            Benjamin W. McInroe, R. Full, D. Koditschek, Yuliy M. Baryshnikov
           </td>
-          <td>2024-08-06</td>
+          <td>2024-10-03</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>2</td>
+          <td>0</td>
+          <td>68</td>
         </tr>
 
-        <tr id="An exponential growth in computing power, which has brought more sophisticated and higher resolution simulations of the climate system, and an exponential increase in observations since the first weather satellite was put in orbit, are revolutionizing climate science. Big data and associated algorithms, coalesced under the field of Machine Learning (ML), offer the opportunity to study the physics of the climate system in ways, and with an amount of detail, infeasible few years ago. The inference provided by ML has allowed to ask causal questions and improve prediction skills beyond classical barriers. Furthermore, when paired with modeling experiments or robust research in model parameterizations, ML is accelerating computations, increasing accuracy and allowing for generating very large ensembles at a fraction of the cost. In light of the urgency imposed by climate change and the rapidly growing role of ML, we review its broader accomplishments in climate physics. Decades long standing problems in observational data reconstruction, representation of sub-grid scale phenomena and climate (and weather) prediction are being tackled with new and justified optimism. Ultimately, this review aims at providing a perspective on the benefits and major challenges of exploiting ML in studying complex systems.">
+        <tr id="Numerical solving parameterised partial differential equations (P-PDEs) is highly practical yet computationally expensive, driving the development of reduced-order models (ROMs). Recently, methods that combine latent space identification techniques with deep learning algorithms (e.g., autoencoders) have shown great potential in describing the dynamical system in the lower dimensional latent space, for example, LaSDI, gLaSDI and GPLaSDI. In this paper, a new parametric latent identification of nonlinear dynamics neural networks, P-TLDINets, is introduced, which relies on a novel neural network structure based on Taylor series expansion and ResNets to learn the ODEs that govern the reduced space dynamics. During the training process, Taylor series-based Latent Dynamic Neural Networks (TLDNets) and identified equations are trained simultaneously to generate a smoother latent space. In order to facilitate the parameterised study, a $k$-nearest neighbours (KNN) method based on an inverse distance weighting (IDW) interpolation scheme is introduced to predict the identified ODE coefficients using local information. Compared to other latent dynamics identification methods based on autoencoders, P-TLDINets remain the interpretability of the model. Additionally, it circumvents the building of explicit autoencoders, avoids dependency on specific grids, and features a more lightweight structure, which is easy to train with high generalisation capability and accuracy. Also, it is capable of using different scales of meshes. P-TLDINets improve training speeds nearly hundred times compared to GPLaSDI and gLaSDI, maintaining an $L_2$ error below $2\%$ compared to high-fidelity models.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/a92dc27ba04348cdf71682bfa77081d8a478b78f" target='_blank'>
-              Machine Learning for the Physics of Climate
+          <td><a href="https://www.semanticscholar.org/paper/808ca10c65d361e7ae56c7ae32591609a48cdf08" target='_blank'>
+              Parametric Taylor series based latent dynamics identification neural networks
               </a>
             </td>
           <td>
-            Annalisa Bracco, Julien Brajard, Henk Dijkstra, P. Hassanzadeh, Christian Lessig, C. Monteleoni
+            Xinlei Lin, Dunhui Xiao
           </td>
-          <td>2024-08-19</td>
+          <td>2024-10-05</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>25</td>
+          <td>0</td>
+          <td>0</td>
         </tr>
 
-        <tr id="In our previous paper [N. Tsutsumi, K. Nakai and Y. Saiki, Chaos 32, 091101 (2022)], we proposed a method for constructing a system of differential equations of chaotic behavior from only observable deterministic time series, which we call the radial function-based regression (RfR) method. However, when the targeted variable's behavior is rather complex, the direct application of the RfR method does not function well. In this study, we propose a novel method of modeling such dynamics, including the high-frequency intermittent behavior of a fluid flow, by considering another variable (base variable) showing relatively simple, less intermittent behavior. We construct an autonomous joint model composed of two parts: the first is an autonomous system of a base variable, and the other concerns the targeted variable being affected by a term involving the base variable to demonstrate complex dynamics. The constructed joint model succeeded in not only inferring a short trajectory but also reconstructing chaotic sets and statistical properties obtained from a long trajectory such as the density distributions of the actual dynamics.">
+        <tr id="The accurate modelling of structural dynamics is crucial across numerous engineering applications, such as Structural Health Monitoring (SHM), seismic analysis, and vibration control. Often, these models originate from physics-based principles and can be derived from corresponding governing equations, often of differential equation form. However, complex system characteristics, such as nonlinearities and energy dissipation mechanisms, often imply that such models are approximative and often imprecise. This challenge is further compounded in SHM, where sensor data is often sparse, making it difficult to fully observe the system's states. To address these issues, this paper explores the use of Physics-Informed Neural Networks (PINNs), a class of physics-enhanced machine learning (PEML) techniques, for the identification and estimation of dynamical systems. PINNs offer a unique advantage by embedding known physical laws directly into the neural network's loss function, allowing for simple embedding of complex phenomena, even in the presence of uncertainties. This study specifically investigates three key applications of PINNs: state estimation in systems with sparse sensing, joint state-parameter estimation, when both system response and parameters are unknown, and parameter estimation within a Bayesian framework to quantify uncertainties. The results demonstrate that PINNs deliver an efficient tool across all aforementioned tasks, even in presence of modelling errors. However, these errors tend to have a more significant impact on parameter estimation, as the optimization process must reconcile discrepancies between the prescribed model and the true system behavior. Despite these challenges, PINNs show promise in dynamical system modeling, offering a robust approach to handling uncertainties.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/1f47a41105c5ebd39e6c06af90c458d59cd20f1b" target='_blank'>
-              Data-driven ODE modeling of the high-frequency complex dynamics of a fluid flow
+          <td><a href="https://www.semanticscholar.org/paper/1fd559dd4ebc4fad5422aba07645bb4cd29ec281" target='_blank'>
+              Response Estimation and System Identification of Dynamical Systems via Physics-Informed Neural Networks
               </a>
             </td>
           <td>
-            Natsuki Tsutsumi, Kengo Nakai, Yoshitaka Saiki
+            M. Haywood-Alexander, Giacamo Arcieri, A. Kamariotis, Eleni Chatzi
           </td>
-          <td>2024-09-01</td>
+          <td>2024-10-02</td>
           <td>ArXiv</td>
           <td>0</td>
           <td>4</td>
         </tr>
 
-        <tr id="The identification of self-similarity is an indispensable tool for understanding and modelling physical phenomena. Unfortunately, this is not always possible to perform formally in highly complex problems. We propose a methodology to extract the similarity variables of a self-similar physical process directly from data, without prior knowledge of the governing equations or boundary conditions, based on an optimization problem and symbolic regression. We analyze the accuracy and robustness of our method in four problems which have been influential in fluid mechanics research: a laminar boundary layer, Burger's equation, a turbulent wake, and a collapsing cavity. Our analysis considers datasets acquired via both numerical and wind-tunnel experiments.">
+        <tr id="In applications, an anticipated situation is where the system of interest has never been encountered before and sparse observations can be made only once. Can the dynamics be faithfully reconstructed from the limited observations without any training data? This problem defies any known traditional methods of nonlinear time-series analysis as well as existing machine-learning methods that typically require extensive data from the target system for training. We address this challenge by developing a hybrid transformer and reservoir-computing machine-learning scheme. The key idea is that, for a complex and nonlinear target system, the training of the transformer can be conducted not using any data from the target system, but with essentially unlimited synthetic data from known chaotic systems. The trained transformer is then tested with the sparse data from the target system. The output of the transformer is further fed into a reservoir computer for predicting the long-term dynamics or the attractor of the target system. The power of the proposed hybrid machine-learning framework is demonstrated using a large number of prototypical nonlinear dynamical systems, with high reconstruction accuracy even when the available data is only 20% of that required to faithfully represent the dynamical behavior of the underlying system. The framework provides a paradigm of reconstructing complex and nonlinear dynamics in the extreme situation where training data does not exist and the observations are random and sparse.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/b4e59a9e878c505fc222908a8279de60c32a29db" target='_blank'>
-              Extracting self-similarity from data
+          <td><a href="https://www.semanticscholar.org/paper/fc970f1344c1636b3bf0229f03757bd2cf4c5445" target='_blank'>
+              Reconstructing dynamics from sparse observations with no training on target system
               </a>
             </td>
           <td>
-            Nikos Bempedelis, Luca Magri, Konstantinos Steiros
+            Zheng-Meng Zhai, Jun-Yin Huang, Benjamin D. Stern, Ying-Cheng Lai
           </td>
-          <td>2024-07-15</td>
+          <td>2024-10-28</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>5</td>
         </tr>
 
-        <tr id="We use parsimonious diffusion maps (PDMs) to discover the latent dynamics of high-fidelity Navier-Stokes simulations with a focus on the 2D fluidic pinball problem. By varying the Reynolds number, different flow regimes emerge, ranging from steady symmetric flows to quasi-periodic asymmetric and turbulence. We show, that the proposed non-linear manifold learning scheme, identifies in a crisp manner the expected intrinsic dimension of the underlying emerging dynamics over the parameter space. In particular, PDMs, estimate that the emergent dynamics in the oscillatory regime can be captured by just two variables, while in the chaotic regime, the dominant modes are three as anticipated by the normal form theory. On the other hand, proper orthogonal decomposition (POD)/PCA, most commonly used for dimensionality reduction in fluid mechanics, does not provide such a crisp separation between the dominant modes. To validate the performance of PDMs, we also computed the reconstruction error, by constructing a decoder using Geometric Harmonics. We show that the proposed scheme outperforms the POD/PCA over the whole Reynolds number range. Thus, we believe that the proposed scheme will allow for the development of more accurate reduced order models for high-fidelity fluid dynamics simulators, thus relaxing the curse of dimensionality in numerical analysis tasks such as bifurcation analysis, optimization and control.">
+        <tr id="The celebrated Takens' embedding theorem provides a theoretical foundation for reconstructing the full state of a dynamical system from partial observations. However, the classical theorem assumes that the underlying system is deterministic and that observations are noise-free, limiting its applicability in real-world scenarios. Motivated by these limitations, we rigorously establish a measure-theoretic generalization that adopts an Eulerian description of the dynamics and recasts the embedding as a pushforward map between probability spaces. Our mathematical results leverage recent advances in optimal transportation theory. Building on our novel measure-theoretic time-delay embedding theory, we have developed a new computational framework that forecasts the full state of a dynamical system from time-lagged partial observations, engineered with better robustness to handle sparse and noisy data. We showcase the efficacy and versatility of our approach through several numerical examples, ranging from the classic Lorenz-63 system to large-scale, real-world applications such as NOAA sea surface temperature forecasting and ERA5 wind field reconstruction.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/62e263f73e87d937004633a6c97b19b9217c4c1e" target='_blank'>
-              Learning the Latent dynamics of Fluid flows from High-Fidelity Numerical Simulations using Parsimonious Diffusion Maps
+          <td><a href="https://www.semanticscholar.org/paper/005c02e2fb7a7bc5cbe28f120e73daaf0c9f2dcc" target='_blank'>
+              Measure-Theoretic Time-Delay Embedding
               </a>
             </td>
           <td>
-            Alessandro Della Pia, Dimitris G. Patsatzis, Lucia Russo, C. Siettos
+            Jonah Botvinick-Greenhouse, Maria Oprea, R. Maulik, Yunan Yang
           </td>
-          <td>2024-08-05</td>
+          <td>2024-09-13</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>24</td>
+          <td>22</td>
+        </tr>
+
+        <tr id="In recent years, live-cell imaging has generated detailed spatiotemporal datasets of biochemical networks within cells. These networks often exhibit characteristics of spatially-distributed excitable systems, with propagating waves of signaling activity that govern processes such as cell migration, division, and other essential physiological functions. Traditionally, these reaction-diffusion systems have been modeled using stochastic partial differential equations incorporating spatial Langevin-type dynamics. Although these knowledge-based models have provided valuable insights, they are typically not directly inferred from experimental data. In this study, we introduce and apply two data-driven methodologies for learning the structure and parameters of spatial Langevin equation (SLE) models: 1) Kramers-Moyal regression, an adaptation of an established approach to fit microscale stochastic dynamics, and 2) wave features optimization, a novel approach to match macroscale spatiotemporal patterns. Unlike black-box neural network models that predict system behavior but provide no mechanistic insight, this approach estimates nonlinear model equations that directly relate to system structure and dynamics. As a proof-of-concept, we focus on simulation datasets derived from two stochastic reaction-diffusion models: one based on the FitzHugh-Nagumo equations (FHN model) and another on a biochemically adapted version of the FHN equations (FR model). Our results demonstrate that optimizing stochastic dynamics and wave patterns enables the accurate estimation of SLE model structure and parameters directly from 1D and 2D spatial data. With sparsity enforcement, we also identify novel sparse reaction-diffusion models for excitable systems. We show that this approach effectively approximates system behavior even when working with datasets similar to experimental conditions, with low temporal resolution and unobserved molecular components. By leveraging machine learning techniques for robust estimation of excitable reaction-diffusion models from spatiotemporal data, this work enhances our ability to model and understand the complex systems that regulate cell behavior.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/2476518d1b9e5bf079d4defed66b90fe1272970f" target='_blank'>
+              Machine learning sparse reaction-diffusion models from stochastic dynamics and spatiotemporal patterns
+              </a>
+            </td>
+          <td>
+            Bedri Abubaker-Sharif, Peter N. Devreotes, Pablo A. Iglesias
+          </td>
+          <td>2024-10-03</td>
+          <td>bioRxiv</td>
+          <td>0</td>
+          <td>4</td>
         </tr>
 
-        <tr id="Steering a system towards a desired target in a very short amount of time is challenging from a computational standpoint. Indeed, the intrinsically iterative nature of optimal control problems requires multiple simulations of the physical system to be controlled. Moreover, the control action needs to be updated whenever the underlying scenario undergoes variations. Full-order models based on, e.g., the Finite Element Method, do not meet these requirements due to the computational burden they usually entail. On the other hand, conventional reduced order modeling techniques such as the Reduced Basis method, are intrusive, rely on a linear superimposition of modes, and lack of efficiency when addressing nonlinear time-dependent dynamics. In this work, we propose a non-intrusive Deep Learning-based Reduced Order Modeling (DL-ROM) technique for the rapid control of systems described in terms of parametrized PDEs in multiple scenarios. In particular, optimal full-order snapshots are generated and properly reduced by either Proper Orthogonal Decomposition or deep autoencoders (or a combination thereof) while feedforward neural networks are exploited to learn the map from scenario parameters to reduced optimal solutions. Nonlinear dimensionality reduction therefore allows us to consider state variables and control actions that are both low-dimensional and distributed. After (i) data generation, (ii) dimensionality reduction, and (iii) neural networks training in the offline phase, optimal control strategies can be rapidly retrieved in an online phase for any scenario of interest. The computational speedup and the high accuracy obtained with the proposed approach are assessed on different PDE-constrained optimization problems, ranging from the minimization of energy dissipation in incompressible flows modelled through Navier-Stokes equations to the thermal active cooling in heat transfer.">
+        <tr id="In climate science, models for global warming and weather prediction face significant challenges due to the limited availability of high-quality data and the difficulty in obtaining it, making data efficiency crucial. In the past few years, Scientific Machine Learning (SciML) models have gained tremendous traction as they can be trained in a data-efficient manner, making them highly suitable for real-world climate applications. Despite this, very little attention has been paid to chaotic climate system modeling utilizing SciML methods. In this paper, we have integrated SciML methods into foundational weather models, where we have enhanced large-scale climate predictions with a physics-informed approach that achieves high accuracy with reduced data. We successfully demonstrate that by combining the interpretability of physical climate models with the computational power of neural networks, SciML models can prove to be a reliable tool for modeling climate. This indicates a shift from the traditional black box-based machine learning modeling of climate systems to physics-informed decision-making, leading to effective climate policy implementation.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/915f70178748c7ef3d789390c868584eaf1f8481" target='_blank'>
-              Real-time optimal control of high-dimensional parametrized systems by deep learning-based reduced order models
+          <td><a href="https://www.semanticscholar.org/paper/355515f809b6db16fa0d5af56232aa416a00376b" target='_blank'>
+              Modeling chaotic Lorenz ODE System using Scientific Machine Learning
               </a>
             </td>
           <td>
-            Matteo Tomasetto, Andrea Manzoni, Francesco Braghin
+            Sameera S Kashyap, R. Dandekar, R. Dandekar, S. Panat
           </td>
-          <td>2024-09-09</td>
+          <td>2024-10-09</td>
           <td>ArXiv</td>
           <td>0</td>
+          <td>3</td>
+        </tr>
+
+        <tr id="None">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/382e068c61d06d6556fc2d6a44e2bd464486a883" target='_blank'>
+              Data-driven discovery of chemotactic migration of bacteria via coordinate-invariant machine learning
+              </a>
+            </td>
+          <td>
+            Y. M. Psarellis, Seungjoon Lee, Tapomoy Bhattacharjee, S. Datta, J. M. Bello-Rivas, Ioannis G Kevrekidis
+          </td>
+          <td>2024-10-24</td>
+          <td>BMC Bioinformatics</td>
           <td>0</td>
+          <td>35</td>
         </tr>
 
-        <tr id="Fitting models to data to obtain distributions of consistent parameter values is important for uncertainty quantification, model comparison, and prediction. Standard Markov Chain Monte Carlo (MCMC) approaches for fitting ordinary differential equations (ODEs) to time-series data involve proposing trial parameter sets, numerically integrating the ODEs forward in time, and accepting or rejecting the trial parameter sets. When the model dynamics depend nonlinearly on the parameters, as is generally the case, trial parameter sets are often rejected, and MCMC approaches become prohibitively computationally costly to converge. Here, we build on methods for numerical continuation and trajectory optimization to introduce an approach in which we use Langevin dynamics in the joint space of variables and parameters to sample models that satisfy constraints on the dynamics. We demonstrate the method by sampling Hopf bifurcations and limit cycles of a model of a biochemical oscillator in a Bayesian framework for parameter estimation, and we obtain more than a hundred fold speedup relative to a leading ensemble MCMC approach that requires numerically integrating the ODEs forward in time. We describe numerical experiments that provide insight into the speedup. The method is general and can be used in any framework for parameter estimation and model selection.">
+        <tr id="Solving partial differential equations (PDEs) is essential in scientific forecasting and fluid dynamics. Traditional approaches often incur expensive computational costs and trade-offs in efficiency and accuracy. Recent deep neural networks improve accuracy but require quality training data. Physics-informed neural networks (PINNs) effectively integrate physical laws, reducing data reliance in limited sample scenarios. A novel machine-learning framework, Chebyshev physics-informed Kolmogorov-Arnold network (ChebPIKAN), is proposed to integrate the robust architectures of Kolmogorov-Arnold networks (KAN) with physical constraints to enhance calculation accuracy of PDEs for fluid mechanics. We explore the fundamentals of KAN, emphasis on the advantages of using the orthogonality of Chebyshev polynomial basis functions in spline fitting, and describe the incorporation of physics-informed loss functions tailored to specific PDEs in fluid dynamics, including Allen-Cahn equation, nonlinear Burgers equation, two-dimensional Helmholtz equations, two-dimensional Kovasznay flow and two-dimensional Navier-Stokes equations. Extensive experiments demonstrate that the proposed ChebPIKAN model significantly outperforms standard KAN architecture in solving various PDEs by embedding essential physical information more effectively. These results indicate that augmenting KAN with physical constraints can not only alleviate overfitting issues of KAN but also improve extrapolation performance. Consequently, this study highlights the potential of ChebPIKAN as a powerful tool in computational fluid dynamics, proposing a path toward fast and reliable predictions in fluid mechanics and beyond.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/d584bb91d158ff005e72c0ae8d5fe55aa4b4e9eb" target='_blank'>
-              Sampling parameters of ordinary differential equations with Langevin dynamics that satisfy constraints
+          <td><a href="https://www.semanticscholar.org/paper/4dd0f46fe2d56b8d6854361349a8968b87261e10" target='_blank'>
+              Physics-informed Kolmogorov-Arnold Network with Chebyshev Polynomials for Fluid Mechanics
               </a>
             </td>
           <td>
-            Chris Chi, J. Weare, Aaron R Dinner
+            Chunyu Guo, Lucheng Sun, Shilong Li, Zelong Yuan, Chao Wang
           </td>
-          <td>2024-08-28</td>
+          <td>2024-11-07</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>20</td>
+          <td>1</td>
         </tr>
 
-        <tr id="Understanding how the collective activity of neural populations relates to computation and ultimately behavior is a key goal in neuroscience. To this end, statistical methods which describe high-dimensional neural time series in terms of low-dimensional latent dynamics have played a fundamental role in characterizing neural systems. Yet, what constitutes a successful method involves two opposing criteria: (1) methods should be expressive enough to capture complex nonlinear dynamics, and (2) they should maintain a notion of interpretability often only warranted by simpler linear models. In this paper, we develop an approach that balances these two objectives: the Gaussian Process Switching Linear Dynamical System (gpSLDS). Our method builds on previous work modeling the latent state evolution via a stochastic differential equation whose nonlinear dynamics are described by a Gaussian process (GP-SDEs). We propose a novel kernel function which enforces smoothly interpolated locally linear dynamics, and therefore expresses flexible -- yet interpretable -- dynamics akin to those of recurrent switching linear dynamical systems (rSLDS). Our approach resolves key limitations of the rSLDS such as artifactual oscillations in dynamics near discrete state boundaries, while also providing posterior uncertainty estimates of the dynamics. To fit our models, we leverage a modified learning objective which improves the estimation accuracy of kernel hyperparameters compared to previous GP-SDE fitting approaches. We apply our method to synthetic data and data recorded in two neuroscience experiments and demonstrate favorable performance in comparison to the rSLDS.">
+        <tr id="In an earlier work by a subset of the present authors, the method of the so-called neural deflation was introduced towards identifying a complete set of functionally independent conservation laws of a nonlinear dynamical system. Here, we extend by a significant step this proposal. Instead of using the explicit knowledge of the underlying equations of motion, we develop the method directly from system trajectories. This is crucial towards enhancing the practical implementation of the method in scenarios where solely data reflecting discrete snapshots of the system are available. We showcase the results of the method and the number of associated conservation laws obtained in a diverse range of examples including 1D and 2D harmonic oscillators, the Toda lattice, the Fermi-Pasta-Ulam-Tsingou lattice and the Calogero-Moser system.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/b38dffa6030cac42cce7564c8fe38875646c5f4f" target='_blank'>
-              Modeling Latent Neural Dynamics with Gaussian Process Switching Linear Dynamical Systems
+          <td><a href="https://www.semanticscholar.org/paper/593010842cfed38b8747fecfbc404d092fa3c26e" target='_blank'>
+              Data-Driven Discovery of Conservation Laws from Trajectories via Neural Deflation
               </a>
             </td>
           <td>
-            Amber Hu, D. Zoltowski, Aditya Nair, David Anderson, Lea Duncker, Scott W. Linderman
+            Shaoxuan Chen, Panayotis Kevrekidis, Hong-Kun Zhang, Wei Zhu
           </td>
-          <td>2024-07-19</td>
+          <td>2024-10-07</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>27</td>
+          <td>1</td>
         </tr>
 
-        <tr id="Discovering Ordinary Differential Equations (ODEs) from trajectory data is a crucial task in AI-driven scientific discovery. Recent methods for symbolic discovery of ODEs primarily rely on fixed training datasets collected a-priori, often leading to suboptimal performance, as observed in our experiments in Figure 1. Inspired by active learning, we explore methods for querying informative trajectory data to evaluate predicted ODEs, where data are obtained by the specified initial conditions of the trajectory. Chaos theory indicates that small changes in the initial conditions of a dynamical system can result in vastly different trajectories, necessitating the maintenance of a large set of initial conditions of the trajectory. To address this challenge, we introduce Active Symbolic Discovery of Ordinary Differential Equations via Phase Portrait Sketching (APPS). Instead of directly selecting individual initial conditions, APPS first identifies an informative region and samples a batch of initial conditions within that region. Compared to traditional active learning methods, APPS eliminates the need for maintaining a large amount of data. Extensive experiments demonstrate that APPS consistently discovers more accurate ODE expressions than baseline methods using passively collected datasets.">
+        <tr id="Dynamic mode decomposition (DMD) is a data-driven method of extracting spatial-temporal coherent modes from complex systems and providing an equation-free architecture to model and predict systems. However, in practical applications, the accuracy of DMD can be limited in extracting dynamical features due to sensor noise in measurements. We develop an adaptive method to constantly update dynamic modes and eigenvalues from noisy measurements arising from discrete systems. Our method is based on the Ensemble Kalman filter owing to its capability of handling time-varying systems and nonlinear observables. Our method can be extended to non-autonomous dynamical systems, accurately recovering short-time eigenvalue-eigenvector pairs and observables. Theoretical analysis shows that the estimation is accurate in long term data misfit. We demonstrate the method on both autonomous and non-autonomous dynamical systems to show its effectiveness.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/3feb52bdba8789f0200eb850968f17327721b32e" target='_blank'>
-              Active Symbolic Discovery of Ordinary Differential Equations via Phase Portrait Sketching
+          <td><a href="https://www.semanticscholar.org/paper/21b2840ec7d705b294997c128d9c3352416c60d4" target='_blank'>
+              Estimate of Koopman modes and eigenvalues with Kalman Filter
               </a>
             </td>
           <td>
-            Nan Jiang, Md. Nasim, Yexiang Xue
+            Ningxin Liu, Shuigen Liu, Xin T. Tong, Lijian Jiang
           </td>
-          <td>2024-09-02</td>
+          <td>2024-09-24</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>2</td>
         </tr>
 
-        <tr id="Forecasting dynamical systems is of importance to numerous real-world applications. When possible, dynamical systems forecasts are constructed based on first-principles-based models such as through the use of differential equations. When these equations are unknown, non-intrusive techniques must be utilized to build predictive models from data alone. Machine learning (ML) methods have recently been used for such tasks. Moreover, ML methods provide the added advantage of significant reductions in time-to-solution for predictions in contrast with first-principle based models. However, many state-of-the-art ML-based methods for forecasting rely on neural networks, which may be expensive to train and necessitate requirements for large amounts of memory. In this work, we propose a quantum mechanics inspired ML modeling strategy for learning nonlinear dynamical systems that provides data-driven forecasts for complex dynamical systems with reduced training time and memory costs. This approach, denoted the quantum reservoir computing technique (QRC), is a hybrid quantum-classical framework employing an ensemble of interconnected small quantum systems via classical linear feedback connections. By mapping the dynamical state to a suitable quantum representation amenable to unitary operations, QRC is able to predict complex nonlinear dynamical systems in a stable and accurate manner. We demonstrate the efficacy of this framework through benchmark forecasts of the NOAA Optimal Interpolation Sea Surface Temperature dataset and compare the performance of QRC to other ML methods.">
+        <tr id="We present a theory of causality in dynamical systems using Koopman operators. Our theory is grounded on a rigorous definition of causal mechanism in dynamical systems given in terms of flow maps. In the Koopman framework, we prove that causal mechanisms manifest as particular flows of observables between function subspaces. While the flow map definition is a clear generalization of the standard definition of causal mechanism given in the structural causal model framework, the flow maps are complicated objects that are not tractable to work with in practice. By contrast, the equivalent Koopman definition lends itself to a straightforward data-driven algorithm that can quantify multivariate causal relations in high-dimensional nonlinear dynamical systems. The coupled Rossler system provides examples and demonstrations throughout our exposition. We also demonstrate the utility of our data-driven Koopman causality measure by identifying causal flow in the Lorenz 96 system. We show that the causal flow identified by our data-driven algorithm agrees with the information flow identified through a perturbation propagation experiment. Our work provides new theoretical insights into causality for nonlinear dynamical systems, as well as a new toolkit for data-driven causal analysis.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/ff3c3b75e08bf25e53956e10e92ec03e5cda30f6" target='_blank'>
-              Higher order quantum reservoir computing for non-intrusive reduced-order models
+          <td><a href="https://www.semanticscholar.org/paper/80c1be5dfc68fefd84654d2b0827a8557cb13612" target='_blank'>
+              Causal Discovery in Nonlinear Dynamical Systems using Koopman Operators
               </a>
             </td>
           <td>
-            Vinamr Jain, R. Maulik
+            Adam Rupe, Derek DeSantis, Craig Bakker, Parvathi Kooloth, Jian Lu
           </td>
-          <td>2024-07-31</td>
+          <td>2024-10-14</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>22</td>
+          <td>2</td>
         </tr>
 
-        <tr id="In this work, we present the novel mathematical framework of latent dynamics models (LDMs) for reduced order modeling of parameterized nonlinear time-dependent PDEs. Our framework casts this latter task as a nonlinear dimensionality reduction problem, while constraining the latent state to evolve accordingly to an (unknown) dynamical system. A time-continuous setting is employed to derive error and stability estimates for the LDM approximation of the full order model (FOM) solution. We analyze the impact of using an explicit Runge-Kutta scheme in the time-discrete setting, resulting in the $\Delta\text{LDM}$ formulation, and further explore the learnable setting, $\Delta\text{LDM}_\theta$, where deep neural networks approximate the discrete LDM components, while providing a bounded approximation error with respect to the FOM. Moreover, we extend the concept of parameterized Neural ODE - recently proposed as a possible way to build data-driven dynamical systems with varying input parameters - to be a convolutional architecture, where the input parameters information is injected by means of an affine modulation mechanism, while designing a convolutional autoencoder neural network able to retain spatial-coherence, thus enhancing interpretability at the latent level. Numerical experiments, including the Burgers' and the advection-reaction-diffusion equations, demonstrate the framework's ability to obtain, in a multi-query context, a time-continuous approximation of the FOM solution, thus being able to query the LDM approximation at any given time instance while retaining a prescribed level of accuracy. Our findings highlight the remarkable potential of the proposed LDMs, representing a mathematically rigorous framework to enhance the accuracy and approximation capabilities of reduced order modeling for time-dependent parameterized PDEs.">
+        <tr id="The reconstruction of physical fields from sparse measurements is pivotal in both scientific research and engineering applications. Traditional methods are increasingly supplemented by deep learning models due to their efficacy in extracting features from data. However, except for the low accuracy on complex physical systems, these models often fail to comply with essential physical constraints, such as governing equations and boundary conditions. To overcome this limitation, we introduce a novel data-driven field reconstruction framework, termed the Physics-aligned Schr\"{o}dinger Bridge (PalSB). This framework leverages a diffusion Schr\"{o}dinger bridge mechanism that is specifically tailored to align with physical constraints. The PalSB approach incorporates a dual-stage training process designed to address both local reconstruction mapping and global physical principles. Additionally, a boundary-aware sampling technique is implemented to ensure adherence to physical boundary conditions. We demonstrate the effectiveness of PalSB through its application to three complex nonlinear systems: cylinder flow from Particle Image Velocimetry experiments, two-dimensional turbulence, and a reaction-diffusion system. The results reveal that PalSB not only achieves higher accuracy but also exhibits enhanced compliance with physical constraints compared to existing methods. This highlights PalSB's capability to generate high-quality representations of intricate physical interactions, showcasing its potential for advancing field reconstruction techniques.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2ab0515da234bcf3eca31c334dd3b5ce0b864f00" target='_blank'>
-              On latent dynamics learning in nonlinear reduced order modeling
+          <td><a href="https://www.semanticscholar.org/paper/27d02977437f60b3287aedfb018f690ee79b8b62" target='_blank'>
+              Physics-aligned Schrödinger bridge
               </a>
             </td>
           <td>
-            N. Farenga, S. Fresca, Simone Brivio, A. Manzoni
+            Zeyu Li, Hongkun Dou, Shen Fang, Wang Han, Yue Deng, Lijun Yang
           </td>
-          <td>2024-08-27</td>
+          <td>2024-09-26</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>11</td>
+          <td>2</td>
         </tr>
 
-        <tr id="Accurately predicting the long-term behavior of chaotic systems is crucial for various applications such as climate modeling. However, achieving such predictions typically requires iterative computations over a dense spatiotemporal grid to account for the unstable nature of chaotic systems, which is expensive and impractical in many real-world situations. An alternative approach to such a full-resolved simulation is using a coarse grid and then correcting its errors through a \textit{closure model}, which approximates the overall information from fine scales not captured in the coarse-grid simulation. Recently, ML approaches have been used for closure modeling, but they typically require a large number of training samples from expensive fully-resolved simulations (FRS). In this work, we prove an even more fundamental limitation, i.e., the standard approach to learning closure models suffers from a large approximation error for generic problems, no matter how large the model is, and it stems from the non-uniqueness of the mapping. We propose an alternative end-to-end learning approach using a physics-informed neural operator (PINO) that overcomes this limitation by not using a closure model or a coarse-grid solver. We first train the PINO model on data from a coarse-grid solver and then fine-tune it with (a small amount of) FRS and physics-based losses on a fine grid. The discretization-free nature of neural operators means that they do not suffer from the restriction of a coarse grid that closure models face, and they can provably approximate the long-term statistics of chaotic systems. In our experiments, our PINO model achieves a 120x speedup compared to FRS with a relative error $\sim 5\%$. In contrast, the closure model coupled with a coarse-grid solver is $58$x slower than PINO while having a much higher error $\sim205\%$ when the closure model is trained on the same FRS dataset.">
+        <tr id="Predicting the response of nonlinear dynamical systems subject to random, broadband excitation is important across a range of scientific disciplines, such as structural dynamics and neuroscience. Building data-driven models requires experimental measurements of the system input and output, but it can be difficult to determine whether inaccuracies in the model stem from modelling errors or noise. This paper presents a novel method to identify the causal component of the input-output data from measurements of a system in the presence of output noise, as a function of frequency, without needing a high fidelity model. An output prediction, calculated using an available model, is optimally combined with noisy measurements of the output to predict the input to the system. The parameters of the algorithm balance the two output signals and are utilised to calculate a nonlinear coherence metric as a measure of causality. This method is applicable to a broad class of nonlinear dynamical systems. There are currently no solutions to this problem in the absence of a complete benchmark model.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/ba56bfb99a735364b70e9f185e5b2a87b7ff7d3f" target='_blank'>
-              Beyond Closure Models: Learning Chaotic-Systems via Physics-Informed Neural Operators
+          <td><a href="https://www.semanticscholar.org/paper/bde736ba268909aa2eed737f5cef86d0890cc5a0" target='_blank'>
+              A method for identifying causality in the response of nonlinear dynamical systems
               </a>
             </td>
           <td>
-            Chuwei Wang, Julius Berner, Zong-Yi Li, Di Zhou, Jiayun Wang, Jane Bae, A. Anandkumar
+            Joseph Massingham, Ole Nielsen, Tore Butlin
           </td>
-          <td>2024-08-09</td>
+          <td>2024-09-26</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>18</td>
+          <td>1</td>
         </tr>
 
-        <tr id="Mechanical systems are known to exhibit complex dynamical behavior from harmonic oscillations to chaotic motion. The dynamics undergo qualitative changes due to changes to internal system parameters like stiffness, and also due to changes to external forcing. Mapping out complete bifurcation diagrams numerically or experimentally is resource-consuming, or even infeasible. This study uses a data-driven approach to investigate how bifurcations can be learned from a few system response measurements. Particularly, the concept of reservoir computing (RC) is employed. As proof of concept, a minimal training dataset under the resource constraint problem of a Duffing oscillator with harmonic external forcing is provided as training data. Our results indicate that the RC not only learns to represent the system dynamics for the trained external forcing, but it also manages to provide qualitatively accurate and robust system response predictions for completely unknown \textit{multi-}parameter regimes outside the training data. Particularly, while being trained solely on regular period-2 cycle dynamics, the proposed framework can correctly predict higher-order periodic and even chaotic dynamics for out-of-distribution forcing signals.">
+        <tr id="Dynamical systems (DS) theory is fundamental for many areas of science and engineering. It can provide deep insights into the behavior of systems evolving in time, as typically described by differential or recursive equations. A common approach to facilitate mathematical tractability and interpretability of DS models involves decomposing nonlinear DS into multiple linear DS separated by switching manifolds, i.e. piecewise linear (PWL) systems. PWL models are popular in engineering and a frequent choice in mathematics for analyzing the topological properties of DS. However, hand-crafting such models is tedious and only possible for very low-dimensional scenarios, while inferring them from data usually gives rise to unnecessarily complex representations with very many linear subregions. Here we introduce Almost-Linear Recurrent Neural Networks (AL-RNNs) which automatically and robustly produce most parsimonious PWL representations of DS from time series data, using as few PWL nonlinearities as possible. AL-RNNs can be efficiently trained with any SOTA algorithm for dynamical systems reconstruction (DSR), and naturally give rise to a symbolic encoding of the underlying DS that provably preserves important topological properties. We show that for the Lorenz and R\"ossler systems, AL-RNNs discover, in a purely data-driven way, the known topologically minimal PWL representations of the corresponding chaotic attractors. We further illustrate on two challenging empirical datasets that interpretable symbolic encodings of the dynamics can be achieved, tremendously facilitating mathematical and computational analysis of the underlying systems.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/dc4149ae4b4ab977eeded61e3696b353391dda2c" target='_blank'>
-              Predicting multi-parametric dynamics of externally forced oscillators using reservoir computing and minimal data
+          <td><a href="https://www.semanticscholar.org/paper/cdc4fd1deba6b38640d46f9dd4a912ffaa31ea25" target='_blank'>
+              Almost-Linear RNNs Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction
               </a>
             </td>
           <td>
-            Manish Yadav, Swati Chauhan, M. Shrimali, M. Stender
+            Manuel Brenner, Christoph Jurgen Hemmer, Z. Monfared, Daniel Durstewitz
           </td>
-          <td>2024-08-27</td>
+          <td>2024-10-18</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>19</td>
+          <td>7</td>
         </tr>
 
-        <tr id="Accurate modeling and prediction of complex physical systems often rely on data assimilation techniques to correct errors inherent in model simulations. Traditional methods like the Ensemble Kalman Filter (EnKF) and its variants as well as the recently developed Ensemble Score Filters (EnSF) face significant challenges when dealing with high-dimensional and nonlinear Bayesian filtering problems with sparse observations, which are ubiquitous in real-world applications. In this paper, we propose a novel data assimilation method, Latent-EnSF, which leverages EnSF with efficient and consistent latent representations of the full states and sparse observations to address the joint challenges of high dimensionlity in states and high sparsity in observations for nonlinear Bayesian filtering. We introduce a coupled Variational Autoencoder (VAE) with two encoders to encode the full states and sparse observations in a consistent way guaranteed by a latent distribution matching and regularization as well as a consistent state reconstruction. With comparison to several methods, we demonstrate the higher accuracy, faster convergence, and higher efficiency of Latent-EnSF for two challenging applications with complex models in shallow water wave propagation and medium-range weather forecasting, for highly sparse observations in both space and time.">
+        <tr id="Many important physical systems can be described as the evolution of a Hamiltonian system, which has the important property of being conservative, that is, energy is conserved throughout the evolution. Physics Informed Neural Networks and in particular Hamiltonian Neural Networks have emerged as a mechanism to incorporate structural inductive bias into the NN model. By ensuring physical invariances are conserved, the models exhibit significantly better sample complexity and out-of-distribution accuracy than standard NNs. Learning the Hamiltonian as a function of its canonical variables, typically position and velocity, from sample observations of the system thus becomes a critical task in system identification and long-term prediction of system behavior. However, to truly preserve the long-run physical conservation properties of Hamiltonian systems, one must use symplectic integrators for a forward pass of the system's simulation. While symplectic schemes have been used in the literature, they are thus far limited to situations when they reduce to explicit algorithms, which include the case of separable Hamiltonians or augmented non-separable Hamiltonians. We extend it to generalized non-separable Hamiltonians, and noting the self-adjoint property of symplectic integrators, we bypass computationally intensive backpropagation through an ODE solver. We show that the method is robust to noise and provides a good approximation of the system Hamiltonian when the state variables are sampled from a noisy observation. In the numerical results, we show the performance of the method concerning Hamiltonian reconstruction and conservation, indicating its particular advantage for non-separable systems.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/0eb62ccf8090693e419be323ad544c6aa5a8c513" target='_blank'>
-              Latent-EnSF: A Latent Ensemble Score Filter for High-Dimensional Data Assimilation with Sparse Observation Data
+          <td><a href="https://www.semanticscholar.org/paper/e92316bb21fc0f2dba7399ed2b4dceedd3feac93" target='_blank'>
+              Learning Generalized Hamiltonians using fully Symplectic Mappings
               </a>
             </td>
           <td>
-            Phillip Si, Peng Chen
+            Harsh Choudhary, Chandan Gupta, Vyacheslav kungrutsev, Melvin Leok, Georgios Korpas
           </td>
-          <td>2024-08-29</td>
+          <td>2024-09-17</td>
           <td>ArXiv</td>
           <td>0</td>
           <td>0</td>
         </tr>
 
-        <tr id="Neural simulators for modeling complex dynamical systems have been extensively studied for various real-world applications, such as weather forecasting, ocean current prediction, and computational fluid dynamics simulation. Although they have demonstrated powerful fitting and predicting, most existing models are only built to learn single-system dynamics. Several advanced researches have considered learning dynamics across environments, which can exploit the potential commonalities among the dynamics across environments and adapt to new environments. However, these methods still are prone to scarcity problems where per-environment data is sparse or limited. Therefore, we propose a novel CoNDP (Context-Informed Neural ODE Processes) to achieve learning system dynamics from sparse observations across environments. It can fully use contextual information of each environment to better capture the intrinsic commonalities across environments and distinguishable differences among environments while modeling uncertainty of system evolution, producing more accurate predictions. Intensive experiments are conducted on five complex dynamical systems in various fields. Results show that the proposed CoNDP can achieve optimal results compared with common neural simulators and state-of-the-art cross-environmental models.">
+        <tr id="Chaotic dynamics, commonly seen in weather systems and fluid turbulence, are characterized by their sensitivity to initial conditions, which makes accurate prediction challenging. Despite its sensitivity to initial perturbations, many chaotic systems observe dissipative behaviors and ergodicity. Therefore, recently various approaches have been proposed to develop data-driven models preserving invariant statistics over long horizons. Although these methods have shown empirical success in reducing instances of unbounded trajectory generation, many of the models are still prone to generating unbounded trajectories, leading to invalid statistics evaluation. In this paper, we propose a novel neural network architecture that simultaneously learns a dissipative dynamics emulator that guarantees to generate bounded trajectories and an energy-like function that governs the dissipative behavior. More specifically, by leveraging control-theoretic ideas, we derive algebraic conditions based on the learned energy-like function that ensure asymptotic convergence to an invariant level set. Using these algebraic conditions, our proposed model enforces dissipativity through a ReLU projection layer, which provides formal trajectory boundedness guarantees. Furthermore, the invariant level set provides an outer estimate for the strange attractor, which is known to be very difficult to characterize due to its complex geometry. We demonstrate the capability of our model in producing bounded long-horizon trajectory forecasts and characterizing the attractor for chaotic dynamical systems including Lorenz 96 and a truncated Kuramoto-Sivashinsky equation.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/40337006a651c2110229b8ae928a5d5fe9a01da1" target='_blank'>
-              Stochastic Neural Simulator for Generalizing Dynamical Systems across Environments
+          <td><a href="https://www.semanticscholar.org/paper/72248bca1ae4617067cbe6521226425ebba7a632" target='_blank'>
+              Learning Chaotic Dynamics with Embedded Dissipativity
               </a>
             </td>
           <td>
-            Liu Jiaqi, Jiaxu Cui, Jiayi Yang, Bo Yang
+            Sunbochen Tang, T. Sapsis, Navid Azizan
           </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
+          <td>2024-10-01</td>
+          <td>ArXiv</td>
           <td>0</td>
-          <td>5</td>
+          <td>39</td>
         </tr>
 
-        <tr id="Uncertainty quantification (UQ) is a crucial but challenging task in many high-dimensional regression or learning problems to increase the confidence of a given predictor. We develop a new data-driven approach for UQ in regression that applies both to classical regression approaches such as the LASSO as well as to neural networks. One of the most notable UQ techniques is the debiased LASSO, which modifies the LASSO to allow for the construction of asymptotic confidence intervals by decomposing the estimation error into a Gaussian and an asymptotically vanishing bias component. However, in real-world problems with finite-dimensional data, the bias term is often too significant to be neglected, resulting in overly narrow confidence intervals. Our work rigorously addresses this issue and derives a data-driven adjustment that corrects the confidence intervals for a large class of predictors by estimating the means and variances of the bias terms from training data, exploiting high-dimensional concentration phenomena. This gives rise to non-asymptotic confidence intervals, which can help avoid overestimating uncertainty in critical applications such as MRI diagnosis. Importantly, our analysis extends beyond sparse regression to data-driven predictors like neural networks, enhancing the reliability of model-based deep learning. Our findings bridge the gap between established theory and the practical applicability of such debiased methods.">
+        <tr id="Efficiently estimating system dynamics from data is essential for minimizing data collection costs and improving model performance. This work addresses the challenge of designing future control inputs to maximize information gain, thereby improving the efficiency of the system identification process. We propose an approach that integrates informative input design into the Dynamic Mode Decomposition with control (DMDc) framework, which is well-suited for high-dimensional systems. By formulating an approximate convex optimization problem that minimizes the trace of the estimation error covariance matrix, we are able to efficiently reduce uncertainty in the model parameters while respecting constraints on the system states and control inputs. This method outperforms traditional techniques like Pseudo-Random Binary Sequences (PRBS) and orthogonal multisines, which do not adapt to the current system model and often gather redundant information. We validate our approach using aircraft and fluid dynamics simulations to demonstrate the practical applicability and effectiveness of our method. Our results show that strategically planning control inputs based on the current model enhances the accuracy of system identification while requiring less data. Furthermore, we provide our implementation and simulation interfaces as an open-source software package, facilitating further research development and use by industry practitioners.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/6b74e7097e6bf5e51f1391656df98c0c04de0d93" target='_blank'>
-              Non-Asymptotic Uncertainty Quantification in High-Dimensional Learning
+          <td><a href="https://www.semanticscholar.org/paper/6127ef757da53faa534ca8a3e197c3eefe94e5c0" target='_blank'>
+              Informative Input Design for Dynamic Mode Decomposition
               </a>
             </td>
           <td>
-            Frederik Hoppe, C. M. Verdun, Hannah Laus, Felix Krahmer, Holger Rauhut
+            Joshua Ott, Mykel J. Kochenderfer, Stephen Boyd
           </td>
-          <td>2024-07-18</td>
+          <td>2024-09-19</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>5</td>
+          <td>17</td>
         </tr>
 
-        <tr id="In this paper, we propose a physics-informed learning-based Koopman modeling approach and present a Koopman-based self-tuning moving horizon estimation design for a class of nonlinear systems. Specifically, we train Koopman operators and two neural networks - the state lifting network and the noise characterization network - using both data and available physical information. The two neural networks account for the nonlinear lifting functions for Koopman modeling and describing system noise distributions, respectively. Accordingly, a stochastic linear Koopman model is established in the lifted space to forecast the dynamic behavior of the nonlinear system. Based on the Koopman model, a self-tuning linear moving horizon estimation (MHE) scheme is developed. The weighting matrices of the MHE design are updated using the pre-trained noise characterization network at each sampling instant. The proposed estimation scheme is computationally efficient because only convex optimization is involved during online implementation, and updating the weighting matrices of the MHE scheme does not require re-training the neural networks. We verify the effectiveness and evaluate the performance of the proposed method via the application to a simulated chemical process.">
+        <tr id="We consider the task of data-driven identification of dynamical systems, specifically for systems whose behavior at large frequencies is non-standard, as encoded by a non-trivial relative degree of the transfer function or, alternatively, a non-trivial index of a corresponding realization as a descriptor system. We develop novel surrogate modeling strategies that allow state-of-the-art rational approximation algorithms (e.g., AAA and vector fitting) to better handle data coming from such systems with non-trivial relative degree. Our contribution is twofold. On one hand, we describe a strategy to build rational surrogate models with prescribed relative degree, with the objective of mirroring the high-frequency behavior of the high-fidelity problem, when known. The surrogate model's desired degree is achieved through constraints on its barycentric coefficients, rather than through ad-hoc modifications of the rational form. On the other hand, we present a degree-identification routine that allows one to estimate the unknown relative degree of a system from low-frequency data. By identifying the degree of the system that generated the data, we can build a surrogate model that, in addition to matching the data well (at low frequencies), has enhanced extrapolation capabilities (at high frequencies). We showcase the effectiveness and robustness of the newly proposed method through a suite of numerical tests.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/7f4e3675450b08e1344396a30181355d5565efe8" target='_blank'>
-              Self-tuning moving horizon estimation of nonlinear systems via physics-informed machine learning Koopman modeling
+          <td><a href="https://www.semanticscholar.org/paper/594e09448a7701a8dc362810a7d28f3f0e6378b6" target='_blank'>
+              Barycentric rational approximation for learning the index of a dynamical system from limited data
               </a>
             </td>
           <td>
-            Mingxue Yan, Minghao Han, A. Law, Xunyuan Yin
+            Davide Pradovera, I. V. Gosea, Jan Heiland
           </td>
-          <td>2024-08-07</td>
+          <td>2024-10-02</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>35</td>
+          <td>12</td>
         </tr>
 
-        <tr id="Neural Ordinary Differential Equations (ODEs) represent a significant advancement at the intersection of machine learning and dynamical systems, offering a continuous-time analog to discrete neural networks. Despite their promise, deploying neural ODEs in practical applications often encounters the challenge of stiffness, a condition where rapid variations in some components of the solution demand prohibitively small time steps for explicit solvers. This work addresses the stiffness issue when employing neural ODEs for model order reduction by introducing a suitable reparametrization in time. The considered map is data-driven and it is induced by the adaptive time-stepping of an implicit solver on a reference solution. We show the map produces a nonstiff system that can be cheaply solved with an explicit time integration scheme. The original, stiff, time dynamic is recovered by means of a map learnt by a neural network that connects the state space to the time reparametrization. We validate our method through extensive experiments, demonstrating improvements in efficiency for the neural ODE inference while maintaining robustness and accuracy. The neural network model also showcases good generalization properties for times beyond the training data.">
+        <tr id="In this paper we use Gaussian processes (kernel methods) to learn mappings between trajectories of distinct differential equations. Our goal is to simplify both the representation and the solution of these equations. We begin by examining the Cole-Hopf transformation, a classical result that converts the nonlinear, viscous Burgers' equation into the linear heat equation. We demonstrate that this transformation can be effectively learned using Gaussian process regression, either from single or from multiple initial conditions of the Burgers equation. We then extend our methodology to discover mappings between initial conditions of a nonlinear partial differential equation (PDE) and a linear PDE, where the exact form of the linear PDE remains unknown and is inferred through Computational Graph Completion (CGC), a generalization of Gaussian Process Regression from approximating single input/output functions to approximating multiple input/output functions that interact within a computational graph. Further, we employ CGC to identify a local transformation from the nonlinear ordinary differential equation (ODE) of the Brusselator to its Poincar\'{e} normal form, capturing the dynamics around a Hopf bifurcation. We conclude by addressing the broader question of whether systematic transformations between nonlinear and linear PDEs can generally exist, suggesting avenues for future research.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/fc642cd2dc03d50f310399a91145bd5a14a7f646" target='_blank'>
-              Neural Ordinary Differential Equations for Model Order Reduction of Stiff Systems
+          <td><a href="https://www.semanticscholar.org/paper/bc191538175874d9f5a15a892a1c5f358da4ed5d" target='_blank'>
+              Gaussian Processes simplify differential equations
               </a>
             </td>
           <td>
-            Matteo Caldana, J. Hesthaven
+            Jonghyeon Lee, B. Hamzi, Y. Kevrekidis, H. Owhadi
           </td>
-          <td>2024-08-12</td>
+          <td>2024-10-03</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>64</td>
+          <td>36</td>
         </tr>
 
-        <tr id="Complex dynamic systems are typically either modeled using expert knowledge in the form of differential equations or via data-driven universal approximation models such as artificial neural networks (ANN). While the first approach has advantages with respect to interpretability, transparency, data-efficiency, and extrapolation, the second approach is able to learn completely unknown functional relations from data and may result in models that can be evaluated more efficiently. To combine the complementary advantages, universal differential equations (UDE) have been suggested. They replace unknown terms in the differential equations with ANN. Such hybrid models allow to both encode prior domain knowledge, such as first principles, and to learn unknown mechanisms from data. Often, data for the training of UDE can only be obtained via costly experiments. We consider optimal experimental design (OED) for planning of experiments and generating data needed to train UDE. The number of weights in the embedded ANN usually leads to an overfitting of the regression problem. To make the OED problem tractable for optimization, we propose and compare dimension reduction methods that are based on lumping of weights and singular value decomposition of the Fisher information matrix (FIM), respectively. They result in lower-dimensional variational differential equations, which are easier to solve and yield regular FIM. Our numerical results showcase the advantages of OED for UDE, such as increased data-efficiency and better extrapolation properties.">
+        <tr id="The fluctuation-dissipation theorem is a cornerstone result in statistical mechanics that can be used to translate the statistics of the free natural variability of a system into information on its forced response to perturbations. By combining this viewpoint on response theory with the key ingredients of Koopmanism, it is possible to deconstruct virtually any response operator into a sum of terms, each associated with a specific mode of natural variability of the system. This dramatically improves the interpretability of the resulting response formulas. We show here on three simple yet mathematically meaningful examples how to use the Extended Dynamical Mode Decomposition (EDMD) algorithm on an individual trajectory of the system to compute with high accuracy correlation functions as well as Green functions associated with acting forcings. This demonstrates the great potential of using Koopman analysis for the key problem of evaluating and testing the sensitivity of a complex system.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2230fd0495ec13469526bd77cb385acbdb4bddfa" target='_blank'>
-              Optimal Experimental Design for Universal Differential Equations
+          <td><a href="https://www.semanticscholar.org/paper/1084301b4c82602dbd6e472d6ae28af986788f13" target='_blank'>
+              Bridging the Gap between Koopmanism and Response Theory: Using Natural Variability to Predict Forced Response
               </a>
             </td>
           <td>
-            Christoph Plate, Carl Julius Martensen, Sebastian Sager
+            Niccolò Zagli, Matthew Colbrook, Valerio Lucarini, Igor Mezi'c, John Moroney
           </td>
-          <td>2024-08-13</td>
+          <td>2024-10-02</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>5</td>
         </tr>
 
-        <tr id="We present a new surrogate model for emulating the behavior of complex nonlinear dynamical systems with external stochastic excitation. The model represents the system dynamics in state space form through a sparse Kriging model. The resulting surrogate model is termed state space Kriging (S2K) model. Sparsity in the Kriging model is achieved by selecting an informative training subset from the observed time histories of the state vector and its derivative with respect to time. We propose a tailored technique for designing the training time histories of state vector and its derivative, aimed at enhancing the robustness of the S2K prediction. We validate the performance of the S2K model with various benchmarks. The results show that S2K yields accurate prediction of complex nonlinear dynamical systems under stochastic excitation with only a few training time histories of state vector.">
+        <tr id="Computationally efficient solutions for pseudometrics quantifying deviation from topological conjugacy between dynamical systems are presented. Deviation from conjugacy is quantified in a Pareto optimal sense that accounts for spectral properties of Koopman operators as well as trajectory geometry. Theoretical justification is provided for computing such pseudometrics in Koopman eigenfunction space rather than observable space. Furthermore, it is shown that theoretical consistency with topological conjugacy can be maintained when restricting the search for optimal transformations between systems to the unitary group. Therefore the pseudometrics are based on analytical solutions for unitary transformations in Koopman eigenfunction space. Geometric considerations for the deviation from conjugacy Pareto optimality problem are used to develop scalar pseudometrics that account for all possible optimal solutions given just two Pareto points. The approach is demonstrated on two example problems; the first being a simple benchmarking problem and the second an engineering example comparing the dynamics of morphological computation of biological nonlinear muscle actuators to simplified mad-made (including bioinspired) approaches. The benefits of considering operator and trajectory geometry based dissimilarity measures in a unified and consistent formalism is demonstrated. Overall, the deviation from conjugacy pseudometrics provide practical advantages in terms of efficiency and scalability, while maintaining theoretical consistency.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/05c477ea05b00f98f2ba51615f625ccbb52fbe0c" target='_blank'>
-              State Space Kriging model for emulating complex nonlinear dynamical systems under stochastic excitation
+          <td><a href="https://www.semanticscholar.org/paper/2d390a87c3f460a091e99066fd8147dcbeb83eaa" target='_blank'>
+              Efficient pseudometrics for data-driven comparisons of nonlinear dynamical systems
               </a>
             </td>
           <td>
-            Kai Chenga, Iason Papaioannoua, MengZe Lyub, Daniel Straub
+            Bryan Glaz
           </td>
-          <td>2024-09-04</td>
+          <td>2024-09-27</td>
           <td>ArXiv</td>
           <td>0</td>
           <td>0</td>
         </tr>
 
-        <tr id="This paper introduces a novel methodology for solving distributed-order fractional differential equations using a physics-informed machine learning framework. The core of this approach involves extending the support vector regression (SVR) algorithm to approximate the unknown solutions of the governing equations during the training phase. By embedding the distributed-order functional equation into the SVR framework, we incorporate physical laws directly into the learning process. To further enhance computational efficiency, Gegenbauer orthogonal polynomials are employed as the kernel function, capitalizing on their fractional differentiation properties to streamline the problem formulation. Finally, the resulting optimization problem of SVR is addressed either as a quadratic programming problem or as a positive definite system in its dual form. The effectiveness of the proposed approach is validated through a series of numerical experiments on Caputo-based distributed-order fractional differential equations, encompassing both ordinary and partial derivatives.">
+        <tr id="Koopman spectral analysis plays a crucial role in understanding and modeling nonlinear dynamical systems as it reveals key system behaviors and long-term dynamics. However, the presence of measurement noise poses a significant challenge to accurately extracting spectral properties. In this work, we propose a robust method for identifying the Koopman operator and extracting its spectral characteristics in noisy environments. To address the impact of noise, our approach tackles an identification problem that accounts for both systematic errors from finite-dimensional approximations and measurement noise in the data. By incorporating Bayesian learning and Kalman smoothing, the method simultaneously identifies the Koopman operator and estimates system states, effectively decoupling these two error sources. The method's efficiency and robustness are demonstrated through extensive experiments, showcasing its accuracy across varying noise levels.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/678e571496c32c6339a102d96524511b6072448e" target='_blank'>
-              A Physics-Informed Machine Learning Approach for Solving Distributed Order Fractional Differential Equations
+          <td><a href="https://www.semanticscholar.org/paper/cb268b2be065a16a965ef8982796602b22751276" target='_blank'>
+              Koopman Spectral Analysis from Noisy Measurements based on Bayesian Learning and Kalman Smoothing
               </a>
             </td>
           <td>
-            A. Aghaei
+            Zhexuan Zeng, Jun Zhou, Yasen Wang, Zuowei Ping
           </td>
-          <td>2024-09-05</td>
+          <td>2024-10-01</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>5</td>
+          <td>0</td>
         </tr>
 
-        <tr id="Nonlinear optimization (NOPT) is a meaningful tool for solving complex tasks in fields like engineering, economics, and operations research, among others. However, NOPT has problems when it comes to dealing with data variability and noisy input measurements that lead to incorrect solutions. Furthermore, nonlinear constraints may result in outcomes that are either infeasible or suboptimal, such as nonconvex optimization. This paper introduces a novel regularized physics-informed neural network (RPINN) framework as a new NOPT tool for both supervised and unsupervised data-driven scenarios. Our RPINN is threefold: By using custom activation functions and regularization penalties in an artificial neural network (ANN), RPINN can handle data variability and noisy inputs. Furthermore, it employs physics principles to construct the network architecture, computing the optimization variables based on network weights and learned features. In addition, it uses automatic differentiation training to make the system scalable and cut down on computation time through batch-based back-propagation. The test results for both supervised and unsupervised NOPT tasks show that our RPINN can provide solutions that are competitive compared to state-of-the-art solvers. In turn, the robustness of RPINN against noisy input measurements makes it particularly valuable in environments with fluctuating information. Specifically, we test a uniform mixture model and a gas-powered system as NOPT scenarios. Overall, with RPINN, its ANN-based foundation offers significant flexibility and scalability.">
+        <tr id="Echo State Networks (ESNs) are recurrent neural networks usually employed for modeling nonlinear dynamic systems with relatively ease of training. By incorporating physical laws into the training of ESNs, Physics-Informed ESNs (PI-ESNs) were proposed initially to model chaotic dynamic systems without external inputs. They require less data for training since Ordinary Differential Equations (ODEs) of the considered system help to regularize the ESN. In this work, the PI-ESN is extended with external inputs to model controllable nonlinear dynamic systems. Additionally, an existing self-adaptive balancing loss method is employed to balance the contributions of the residual regression term and the physics-informed loss term in the total loss function. The experiments with two nonlinear systems modeled by ODEs, the Van der Pol oscillator and the four-tank system, and with one differential-algebraic (DAE) system, an electric submersible pump, revealed that the proposed PI-ESN outperforms the conventional ESN, especially in scenarios with limited data availability, showing that PI-ESNs can regularize an ESN model with external inputs previously trained on just a few datapoints, reducing its overfitting and improving its generalization error (up to 92% relative reduction in the test error). Further experiments demonstrated that the proposed PI-ESN is robust to parametric uncertainties in the ODE equations and that model predictive control using PI-ESN outperforms the one using plain ESN, particularly when training data is scarce.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/516f48e222fdad35b92113ecdbf8a459d37c95aa" target='_blank'>
-              A Regularized Physics-Informed Neural Network to Support Data-Driven Nonlinear Constrained Optimization
+          <td><a href="https://www.semanticscholar.org/paper/e0b88bf05b77c55de080e84b7fc70b5d74109970" target='_blank'>
+              Physics-Informed Echo State Networks for Modeling Controllable Dynamical Systems
               </a>
             </td>
           <td>
-            Diego Armando Perez-Rosero, A. Álvarez-Meza, G. Castellanos-Domínguez
+            Eric Mochiutti, Eric A. Antonelo, Eduardo Camponogara
           </td>
-          <td>2024-07-18</td>
-          <td>Comput.</td>
+          <td>2024-09-27</td>
+          <td>ArXiv</td>
           <td>0</td>
-          <td>25</td>
+          <td>14</td>
         </tr>
 
-        <tr id="The modeling of dynamical systems is a pervasive concern for not only describing but also predicting and controlling natural phenomena and engineered systems. Current data-driven approaches often assume prior knowledge of the relevant state variables or result in overparameterized state spaces. Boyuan Chen and his co-authors proposed a neural network model that estimates the degrees of freedom and attempts to discover the state variables of a dynamical system. Despite its innovative approach, this baseline model lacks a connection to the physical principles governing the systems it analyzes, leading to unreliable state variables. This research proposes a method that leverages the physical characteristics of second-order Hamiltonian systems to constrain the baseline model. The proposed model outperforms the baseline model in identifying a minimal set of non-redundant and interpretable state variables.">
+        <tr id="Physics-Informed Neural Networks (PINNs) are used to study vortex solutions in the 2+1 dimensional nonlinear partial differential equations. These solutions include the regularized long-wave (RLW) equation and the Zakharov-Kuznetsov (ZK) equation, which are toy models of the geostrophic shallow water model in the planetary atmosphere. PINNs successfully solve these equations in the forward process and the solutions are obtained using the mesh-free approach and automatic differentiation while accounting for conservation laws. In the inverse process, the proper equations can be successfully derived from a given training data. Since these equations have a lot in common, there are situations when substantial misidentification arises during the inverse analysis. We consider PINNs with conservation laws (referred to as cPINNs), deformations of the initial profiles, and a friction approach that provides excellent discrimination of the equations to improve the identification's resolution.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/ddabca2c86de837436088e84f6e03ca06d2c3870" target='_blank'>
-              Physics-informed Discovery of State Variables in Second-Order and Hamiltonian Systems
+          <td><a href="https://www.semanticscholar.org/paper/09235f2156a7d20de3910de1509ae6d574bde0a5" target='_blank'>
+              Discovery of Quasi-Integrable Equations from traveling-wave data using the Physics-Informed Neural Networks
               </a>
             </td>
           <td>
-            Félix Chavelli, Zi-Yu Khoo, Dawen Wu, Jonathan Sze Choong Low, Stéphane Bressan
+            A. Nakamula, N. Sawado, K. Shimasaki, Y. Shimazaki, Y. Suzuki, K. Toda
           </td>
-          <td>2024-08-21</td>
+          <td>2024-10-23</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>3</td>
+          <td>1</td>
         </tr>
 
-        <tr id="Lifted linear predictor (LLP) is an artificial linear dynamical system designed to predict trajectories of a generally nonlinear dynamical system based on the current state (or measurements) and the input. The main benefit of the LLP is its potential ability to capture the nonlinear system's dynamics with precision superior to other linearization techniques, such as local linearization about the operation point. The idea of lifting is supported by the theory of Koopman Operators. For LLP identification, we focus on the data-driven method based on the extended dynamic mode decomposition (EDMD) algorithm. However, while the EDMD algorithm presents an extremely simple and efficient way to obtain the LLP, it can also yield poor results. In this paper, we present some less intuitive practical guidelines for data-driven identification of the LLPs, aiming at improving usability of LLPs for designing control. We support the guidelines with two motivating examples. The implementation of the examples are shared on a public repository.">
+        <tr id="Deep learning has achieved great success in modeling dynamical systems, providing data-driven simulators to predict complex phenomena, even without known governing equations. However, existing models have two major limitations: their narrow focus on mechanical systems and their tendency to treat systems as monolithic. These limitations reduce their applicability to dynamical systems in other domains, such as electrical and hydraulic systems, and to coupled systems. To address these limitations, we propose Poisson-Dirac Neural Networks (PoDiNNs), a novel framework based on the Dirac structure that unifies the port-Hamiltonian and Poisson formulations from geometric mechanics. This framework enables a unified representation of various dynamical systems across multiple domains as well as their interactions and degeneracies arising from couplings. Our experiments demonstrate that PoDiNNs offer improved accuracy and interpretability in modeling unknown coupled dynamical systems from data.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/bf1ce9a8c8346f1264e7cecb5aea44d98aa65af7" target='_blank'>
-              Practical Guidelines for Data-driven Identification of Lifted Linear Predictors for Control
+          <td><a href="https://www.semanticscholar.org/paper/151acdc6b2f8dc3711777ee98c45ff7a1c2e7507" target='_blank'>
+              Poisson-Dirac Neural Networks for Modeling Coupled Dynamical Systems across Domains
               </a>
             </td>
           <td>
-            Loi Do, Adam Uchytil, Zdenvek Hur'ak
+            Razmik Arman Khosrovian, Takaharu Yaguchi, Hiroaki Yoshimura, Takashi Matsubara
           </td>
-          <td>2024-08-02</td>
+          <td>2024-10-15</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>2</td>
+          <td>0</td>
+          <td>11</td>
         </tr>
 
     </tbody>
@@ -1540,7 +1570,7 @@ <h3 id="recommended_articles">Recommended articles on <i>Symbolic regression</i>
 
   var trace1 = {
     x: ['2024'],
-    y: [4],
+    y: [3],
     name: 'Num of citations',
     yaxis: 'y1',
     type: 'scatter'
diff --git a/site/Time-series forecasting/index.html b/site/Time-series forecasting/index.html
index da8eea0f..589127a8 100644
--- a/site/Time-series forecasting/index.html	
+++ b/site/Time-series forecasting/index.html	
@@ -453,7 +453,7 @@ <h1>Time-series forecasting</h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:43 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:40 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -517,8 +517,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
             Ming Jin, Huan Yee Koh, Qingsong Wen, Daniele Zambon, C. Alippi, G. I. Webb, Irwin King, Shirui Pan
           </td>
           <td>2023-07-07</td>
-          <td>IEEE transactions on pattern analysis and machine intelligence</td>
-          <td>64</td>
+          <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
+          <td>82</td>
           <td>50</td>
 
             <td><a href='../recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9' target='_blank'>
@@ -538,7 +538,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2021-10-11</td>
           <td>ArXiv, International Conference on Learning Representations</td>
-          <td>73</td>
+          <td>86</td>
           <td>47</td>
 
             <td><a href='../recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba' target='_blank'>
@@ -558,7 +558,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2023-02-08</td>
           <td>ArXiv, Neural Information Processing Systems</td>
-          <td>18</td>
+          <td>22</td>
           <td>50</td>
 
             <td><a href='../recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419' target='_blank'>
@@ -578,7 +578,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2022-05-26</td>
           <td>J. Mach. Learn. Res., Journal of machine learning research</td>
-          <td>11</td>
+          <td>14</td>
           <td>50</td>
 
             <td><a href='../recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b' target='_blank'>
@@ -598,7 +598,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2023-10-24</td>
           <td>ArXiv, arXiv.org</td>
-          <td>6</td>
+          <td>9</td>
           <td>50</td>
 
             <td><a href='../recommendations/ccea298edb788edf821aef58f0952c3e8debc25a' target='_blank'>
@@ -618,7 +618,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2023-10-11</td>
           <td>ArXiv, Neural Information Processing Systems</td>
-          <td>129</td>
+          <td>198</td>
           <td>14</td>
 
             <td><a href='../recommendations/123acfbccca0460171b6b06a4012dbb991cde55b' target='_blank'>
@@ -638,7 +638,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2024-02-01</td>
           <td>ArXiv, arXiv.org</td>
-          <td>50</td>
+          <td>56</td>
           <td>5</td>
 
             <td><a href='../recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f' target='_blank'>
@@ -658,7 +658,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2023-10-14</td>
           <td>ArXiv, International Conference on Machine Learning</td>
-          <td>51</td>
+          <td>103</td>
           <td>14</td>
 
             <td><a href='../recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f' target='_blank'>
@@ -678,8 +678,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, International Conference on Machine Learning</td>
-          <td>32</td>
-          <td>22</td>
+          <td>71</td>
+          <td>25</td>
 
             <td><a href='../recommendations/4a111f7a3b56d0468f13104999844885157ef17d' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -698,7 +698,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2023-10-03</td>
           <td>ArXiv, International Conference on Learning Representations</td>
-          <td>134</td>
+          <td>195</td>
           <td>9</td>
 
             <td><a href='../recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277' target='_blank'>
@@ -707,7 +707,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
 
         </tr>
 
-        <tr id="Large pre-trained models excel in zero/few-shot learning for language and vision tasks but face challenges in multivariate time series (TS) forecasting due to diverse data characteristics. Consequently, recent research efforts have focused on developing pre-trained TS forecasting models. These models, whether built from scratch or adapted from large language models (LLMs), excel in zero/few-shot forecasting tasks. However, they are limited by slow performance, high computational demands, and neglect of cross-channel and exogenous correlations. To address this, we introduce Tiny Time Mixers (TTM), a compact model (starting from 1M parameters) with effective transfer learning capabilities, trained exclusively on public TS datasets. TTM, based on the light-weight TSMixer architecture, incorporates innovations like adaptive patching, diverse resolution sampling, and resolution prefix tuning to handle pre-training on varied dataset resolutions with minimal model capacity. Additionally, it employs multi-level modeling to capture channel correlations and infuse exogenous signals during fine-tuning. TTM outperforms existing popular benchmarks in zero/few-shot forecasting by (4-40\%), while reducing computational requirements significantly. Moreover, TTMs are lightweight and can be executed even on CPU-only machines, enhancing usability and fostering wider adoption in resource-constrained environments. Model weights for our initial variant (TTM-Q) are available at https://huggingface.co/ibm-granite/granite-timeseries-ttm-v1. Model weights for more sophisticated variants (TTM-B, TTM-E, and TTM-A) will be shared soon. The source code for TTM can be accessed at https://github.com/ibm-granite/granite-tsfm/tree/main/tsfm_public/models/tinytimemixer.">
+        <tr id="Large pre-trained models excel in zero/few-shot learning for language and vision tasks but face challenges in multivariate time series (TS) forecasting due to diverse data characteristics. Consequently, recent research efforts have focused on developing pre-trained TS forecasting models. These models, whether built from scratch or adapted from large language models (LLMs), excel in zero/few-shot forecasting tasks. However, they are limited by slow performance, high computational demands, and neglect of cross-channel and exogenous correlations. To address this, we introduce Tiny Time Mixers (TTM), a compact model (starting from 1M parameters) with effective transfer learning capabilities, trained exclusively on public TS datasets. TTM, based on the light-weight TSMixer architecture, incorporates innovations like adaptive patching, diverse resolution sampling, and resolution prefix tuning to handle pre-training on varied dataset resolutions with minimal model capacity. Additionally, it employs multi-level modeling to capture channel correlations and infuse exogenous signals during fine-tuning. TTM outperforms existing popular benchmarks in zero/few-shot forecasting by (4-40%), while reducing computational requirements significantly. Moreover, TTMs are lightweight and can be executed even on CPU-only machines, enhancing usability and fostering wider adoption in resource-constrained environments. The model weights for reproducibility and research use are available at https://huggingface.co/ibm/ttm-research-r2/, while enterprise-use weights under the Apache license can be accessed as follows: the initial TTM-Q variant at https://huggingface.co/ibm-granite/granite-timeseries-ttm-r1, and the latest variants (TTM-B, TTM-E, TTM-A) weights are available at https://huggingface.co/ibm-granite/granite-timeseries-ttm-r2.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6" target='_blank'>
                 Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series
@@ -718,8 +718,8 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2024-01-08</td>
           <td>ArXiv, arXiv.org</td>
-          <td>2</td>
-          <td>2</td>
+          <td>3</td>
+          <td>3</td>
 
             <td><a href='../recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6' target='_blank'>
               <i class="material-icons">open_in_new</i></a>
@@ -738,7 +738,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2022-06-17</td>
           <td>ArXiv, Neural Information Processing Systems</td>
-          <td>176</td>
+          <td>211</td>
           <td>47</td>
 
             <td><a href='../recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37' target='_blank'>
@@ -758,7 +758,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2023-02-06</td>
           <td>ArXiv, DBLP</td>
-          <td>26</td>
+          <td>38</td>
           <td>47</td>
 
             <td><a href='../recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14' target='_blank'>
@@ -798,7 +798,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2023-01-04</td>
           <td>ArXiv, arXiv.org</td>
-          <td>3</td>
+          <td>4</td>
           <td>50</td>
 
             <td><a href='../recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68' target='_blank'>
@@ -818,7 +818,7 @@ <h3 id="manually_curated_articles">Manually curated articles on <i>Time-series f
           </td>
           <td>2024-02-29</td>
           <td>ArXiv</td>
-          <td>2</td>
+          <td>4</td>
           <td>47</td>
 
             <td><a href='../recommendations/bcbcc2e1af8bcf6b07edf866be95116a8ed0bf91' target='_blank'>
@@ -862,694 +862,664 @@ <h3 id="recommended_articles">Recommended articles on <i>Time-series forecasting
     </thead>
     <tbody>
 
-        <tr id="Time series forecasting (TSF) plays a pivotal role in many real-world applications. Recently, the utilization of Large Language Models (LLM) in TSF has demonstrated exceptional predictive performance, surpassing most task-specific forecasting models. The success of LLM-based forecasting methods underscores the importance of causal dependence modeling and pre-trained knowledge transfer. However, challenges persist in directly applying LLM to TSF, i.e., the unacceptable parameter scales for resource-intensive model optimization, and the significant gap of feature space between structural numerical time series and natural language. To this end, we propose LeRet, a Language-empowered Retentive network for TSF. Technically, inspired by the causal extraction in LLM, we propose a causal dependence learner, enhanced by a patch-level pre-training task, to capture sequential causal evolution. To minimize the gap between numeric and language, we initialize a language description protocol for time series and design a TS-related language knowledge extractor to learn from language description, avoiding training with large-scale parameters. Finally, we dedicatedly achieve a Language-TS Modality Integrator for the fusion of two types data, and enable language-empowered sequence forecasting. Extensive evaluations demonstrate the effectiveness of our LeRet, especially reveal superiority on few-shot, and zero-shot forecasting tasks. Code is available at https://github.com/hqh0728/LeRet.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/a08b1fc4c1621dc54a57b39a4eb7f9f19abbac63" target='_blank'>
-              LeRet: Language-Empowered Retentive Network for Time Series Forecasting
-              </a>
-            </td>
-          <td>
-            Qihe Huang, Zhen-Qiang Zhou, Kuo Yang, Gengyu Lin, Zhongchao Yi, Yang Wang
-          </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
-          <td>0</td>
-          <td>3</td>
-        </tr>
-
-        <tr id="With recent advances in building foundation models for texts and video data, there is a surge of interest in foundation models for time series. A family of models have been developed, utilizing a temporal auto-regressive generative Transformer architecture, whose effectiveness has been proven in Large Language Models. While the empirical results are promising, almost all existing time series foundation models have only been tested on well-curated ``benchmark'' datasets very similar to texts. However, real-world time series exhibit unique challenges, such as variable channel sizes across domains, missing values, and varying signal sampling intervals due to the multi-resolution nature of real-world data. Additionally, the uni-directional nature of temporally auto-regressive decoding limits the incorporation of domain knowledge, such as physical laws expressed as partial differential equations (PDEs). To address these challenges, we introduce the Time Diffusion Transformer (TimeDiT), a general foundation model for time series that employs a denoising diffusion paradigm instead of temporal auto-regressive generation. TimeDiT leverages the Transformer architecture to capture temporal dependencies and employs diffusion processes to generate high-quality candidate samples without imposing stringent assumptions on the target distribution via novel masking schemes and a channel alignment strategy. Furthermore, we propose a finetuning-free model editing strategy that allows the seamless integration of external knowledge during the sampling process without updating any model parameters. Extensive experiments conducted on a varity of tasks such as forecasting, imputation, and anomaly detection, demonstrate the effectiveness of TimeDiT.">
+        <tr id="Time series foundation models have demonstrated strong performance in zero-shot learning, making them well-suited for predicting rapidly evolving patterns in real-world applications where relevant training data are scarce. However, most of these models rely on the Transformer architecture, which incurs quadratic complexity as input length increases. To address this, we introduce TSMamba, a linear-complexity foundation model for time series forecasting built on the Mamba architecture. The model captures temporal dependencies through both forward and backward Mamba encoders, achieving high prediction accuracy. To reduce reliance on large datasets and lower training costs, TSMamba employs a two-stage transfer learning process that leverages pretrained Mamba LLMs, allowing effective time series modeling with a moderate training set. In the first stage, the forward and backward backbones are optimized via patch-wise autoregressive prediction; in the second stage, the model trains a prediction head and refines other components for long-term forecasting. While the backbone assumes channel independence to manage varying channel numbers across datasets, a channel-wise compressed attention module is introduced to capture cross-channel dependencies during fine-tuning on specific multivariate datasets. Experiments show that TSMamba's zero-shot performance is comparable to state-of-the-art time series foundation models, despite using significantly less training data. It also achieves competitive or superior full-shot performance compared to task-specific prediction models. The code will be made publicly available.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/ace7d07de33a903e1b3052479cacd2dfa7ae2bb6" target='_blank'>
-              TimeDiT: General-purpose Diffusion Transformers for Time Series Foundation Model
+          <td><a href="https://www.semanticscholar.org/paper/30abc3fad3c88886c189e216c8414b18c396c6bd" target='_blank'>
+              A Mamba Foundation Model for Time Series Forecasting
               </a>
             </td>
           <td>
-            Defu Cao, Wen Ye, Yizhou Zhang, Yan Liu
+            Haoyu Ma, Yushu Chen, Wenlai Zhao, Jinzhe Yang, Yingsheng Ji, Xinghua Xu, Xiaozhu Liu, Hao Jing, Shengzhuo Liu, Guangwen Yang
           </td>
-          <td>2024-09-03</td>
+          <td>2024-11-05</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>9</td>
+          <td>7</td>
         </tr>
 
-        <tr id="The self-attention mechanism in Transformer architecture, invariant to sequence order, necessitates positional embeddings to encode temporal order in time series prediction. We argue that this reliance on positional embeddings restricts the Transformer's ability to effectively represent temporal sequences, particularly when employing longer lookback windows. To address this, we introduce an innovative approach that combines Pyramid RNN embeddings(PRE) for univariate time series with the Transformer's capability to model multivariate dependencies. PRE, utilizing pyramidal one-dimensional convolutional layers, constructs multiscale convolutional features that preserve temporal order. Additionally, RNNs, layered atop these features, learn multiscale time series representations sensitive to sequence order. This integration into Transformer models with attention mechanisms results in significant performance enhancements. We present the PRformer, a model integrating PRE with a standard Transformer encoder, demonstrating state-of-the-art performance on various real-world datasets. This performance highlights the effectiveness of our approach in leveraging longer lookback windows and underscores the critical role of robust temporal representations in maximizing Transformer's potential for prediction tasks. Code is available at this repository: \url{https://github.com/usualheart/PRformer}.">
+        <tr id="Time series forecasting is prevalent in extensive real-world applications, such as financial analysis and energy planning. Previous studies primarily focus on time series modality, endeavoring to capture the intricate variations and dependencies inherent in time series. Beyond numerical time series data, we notice that metadata (e.g.~dataset and variate descriptions) also carries valuable information essential for forecasting, which can be used to identify the application scenario and provide more interpretable knowledge than digit sequences. Inspired by this observation, we propose a Metadata-informed Time Series Transformer (MetaTST), which incorporates multiple levels of context-specific metadata into Transformer forecasting models to enable informative time series forecasting. To tackle the unstructured nature of metadata, MetaTST formalizes them into natural languages by pre-designed templates and leverages large language models (LLMs) to encode these texts into metadata tokens as a supplement to classic series tokens, resulting in an informative embedding. Further, a Transformer encoder is employed to communicate series and metadata tokens, which can extend series representations by metadata information for more accurate forecasting. This design also allows the model to adaptively learn context-specific patterns across various scenarios, which is particularly effective in handling large-scale, diverse-scenario forecasting tasks. Experimentally, MetaTST achieves state-of-the-art compared to advanced time series models and LLM-based methods on widely acknowledged short- and long-term forecasting benchmarks, covering both single-dataset individual and multi-dataset joint training settings.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/3242d2431167e373a8e2d3ad6a5cc7f3d7c7741e" target='_blank'>
-              PRformer: Pyramidal Recurrent Transformer for Multivariate Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/f46d685e05e6742f73946ce70533672b02e5f1ed" target='_blank'>
+              Metadata Matters for Time Series: Informative Forecasting with Transformers
               </a>
             </td>
           <td>
-            Yongbo Yu, Weizhong Yu, Feiping Nie, Xuelong Li
+            Jiaxiang Dong, Haixu Wu, Yuxuan Wang, Li Zhang, Jianmin Wang, Mingsheng Long
           </td>
-          <td>2024-08-20</td>
+          <td>2024-10-04</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>10</td>
+          <td>67</td>
         </tr>
 
-        <tr id="Long-term time series forecasting in centralized environments poses unique challenges regarding data privacy, communication overhead, and scalability. To address these challenges, we propose FedTime, a federated large language model (LLM) tailored for long-range time series prediction. Specifically, we introduce a federated pre-trained LLM with fine-tuning and alignment strategies. Prior to the learning process, we employ K-means clustering to partition edge devices or clients into distinct clusters, thereby facilitating more focused model training. We also incorporate channel independence and patching to better preserve local semantic information, ensuring that important contextual details are retained while minimizing the risk of information loss. We demonstrate the effectiveness of our FedTime model through extensive experiments on various real-world forecasting benchmarks, showcasing substantial improvements over recent approaches. In addition, we demonstrate the efficiency of FedTime in streamlining resource usage, resulting in reduced communication overhead.">
+        <tr id="Developing a foundation model for time series forecasting across diverse domains has attracted significant attention in recent years. Existing works typically assume regularly sampled, well-structured data, limiting their applicability to more generalized scenarios where time series often contain missing values, unequal sequence lengths, and irregular time intervals between measurements. To cover diverse domains and handle variable regularities, we propose FlexTSF, a universal time series forecasting model that possesses better generalization and natively support both regular and irregular time series. FlexTSF produces forecasts in an autoregressive manner and incorporates three novel designs: VT-Norm, a normalization strategy to ablate data domain barriers, IVP Patcher, a patching module to learn representations from flexibly structured time series, and LED attention, an attention mechanism to seamlessly integrate these two and propagate forecasts with awareness of domain and time information. Experiments on 12 datasets show that FlexTSF outperforms state-of-the-art forecasting models respectively designed for regular and irregular time series. Furthermore, after self-supervised pre-training, FlexTSF shows exceptional performance in both zero-shot and few-show settings for time series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/cfbf583c88970300ba4a74619b02be88524899df" target='_blank'>
-              A federated large language model for long-term time series forecasting
+          <td><a href="https://www.semanticscholar.org/paper/45521591fd8c46068f0a8712b43ce6cdaac1edea" target='_blank'>
+              FlexTSF: A Universal Forecasting Model for Time Series with Variable Regularities
               </a>
             </td>
           <td>
-            Raed Abdel Sater, A. B. Hamza
+            Jingge Xiao, Yile Chen, Gao Cong, Wolfgang Nejdl, Simon Gottschalk
           </td>
-          <td>2024-07-30</td>
+          <td>2024-10-30</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>6</td>
         </tr>
 
-        <tr id="Time series, characterized by a sequence of data points arranged in a discrete-time order, are ubiquitous in real-world applications. Different from other modalities, time series present unique challenges due to their complex and dynamic nature, including the entanglement of nonlinear patterns and time-variant trends. Analyzing time series data is of great significance in real-world scenarios and has been widely studied over centuries. Recent years have witnessed remarkable breakthroughs in the time series community, with techniques shifting from traditional statistical methods to advanced deep learning models. In this paper, we delve into the design of deep time series models across various analysis tasks and review the existing literature from two perspectives: basic modules and model architectures. Further, we develop and release Time Series Library (TSLib) as a fair benchmark of deep time series models for diverse analysis tasks, which implements 24 mainstream models, covers 30 datasets from different domains, and supports five prevalent analysis tasks. Based on TSLib, we thoroughly evaluate 12 advanced deep time series models on different tasks. Empirical results indicate that models with specific structures are well-suited for distinct analytical tasks, which offers insights for research and adoption of deep time series models. Code is available at https://github.com/thuml/Time-Series-Library.">
+        <tr id="Deep learning for time series forecasting has seen significant advancements over the past decades. However, despite the success of large-scale pre-training in language and vision domains, pre-trained time series models remain limited in scale and operate at a high cost, hindering the development of larger capable forecasting models in real-world applications. In response, we introduce Time-MoE, a scalable and unified architecture designed to pre-train larger, more capable forecasting foundation models while reducing inference costs. By leveraging a sparse mixture-of-experts (MoE) design, Time-MoE enhances computational efficiency by activating only a subset of networks for each prediction, reducing computational load while maintaining high model capacity. This allows Time-MoE to scale effectively without a corresponding increase in inference costs. Time-MoE comprises a family of decoder-only transformer models that operate in an auto-regressive manner and support flexible forecasting horizons with varying input context lengths. We pre-trained these models on our newly introduced large-scale data Time-300B, which spans over 9 domains and encompassing over 300 billion time points. For the first time, we scaled a time series foundation model up to 2.4 billion parameters, achieving significantly improved forecasting precision. Our results validate the applicability of scaling laws for training tokens and model size in the context of time series forecasting. Compared to dense models with the same number of activated parameters or equivalent computation budgets, our models consistently outperform them by large margin. These advancements position Time-MoE as a state-of-the-art solution for tackling real-world time series forecasting challenges with superior capability, efficiency, and flexibility.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/425182f5c96c1d239da5cbe3a24371b2bab6319b" target='_blank'>
-              Deep Time Series Models: A Comprehensive Survey and Benchmark
+          <td><a href="https://www.semanticscholar.org/paper/2cb3044ef42c7ee022a988864028b80ce977072c" target='_blank'>
+              Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts
               </a>
             </td>
           <td>
-            Yuxuan Wang, Haixu Wu, Jiaxiang Dong, Yong Liu, Mingsheng Long, Jianmin Wang
+            X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin
           </td>
-          <td>2024-07-18</td>
+          <td>2024-09-24</td>
           <td>ArXiv</td>
-          <td>3</td>
-          <td>66</td>
+          <td>2</td>
+          <td>9</td>
         </tr>
 
-        <tr id="Recently many deep learning models have been proposed for Long-term Time Series Forecasting (LTSF). Based on previous literature, we identify three critical patterns that can improve forecasting accuracy: the order and semantic dependencies in time dimension as well as cross-variate dependency. However, little effort has been made to simultaneously consider order and semantic dependencies when developing forecasting models. Moreover, existing approaches utilize cross-variate dependency by mixing information from different timestamps and variates, which may introduce irrelevant or harmful cross-variate information to the time dimension and largely hinder forecasting performance. To overcome these limitations, we investigate the potential of Mamba for LTSF and discover two key advantages benefiting forecasting: (i) the selection mechanism makes Mamba focus on or ignore specific inputs and learn semantic dependency easily, and (ii) Mamba preserves order dependency by processing sequences recursively. After that, we empirically find that the non-linear activation used in Mamba is unnecessary for semantically sparse time series data. Therefore, we further propose SAMBA, a Simplified Mamba with disentangled dependency encoding. Specifically, we first remove the non-linearities of Mamba to make it more suitable for LTSF. Furthermore, we propose a disentangled dependency encoding strategy to endow Mamba with cross-variate dependency modeling capabilities while reducing the interference between time and variate dimensions. Extensive experimental results on seven real-world datasets demonstrate the effectiveness of SAMBA over state-of-the-art forecasting models.">
+        <tr id="This paper introduces Mamba4Cast, a zero-shot foundation model for time series forecasting. Based on the Mamba architecture and inspired by Prior-data Fitted Networks (PFNs), Mamba4Cast generalizes robustly across diverse time series tasks without the need for dataset specific fine-tuning. Mamba4Cast's key innovation lies in its ability to achieve strong zero-shot performance on real-world datasets while having much lower inference times than time series foundation models based on the transformer architecture. Trained solely on synthetic data, the model generates forecasts for entire horizons in a single pass, outpacing traditional auto-regressive approaches. Our experiments show that Mamba4Cast performs competitively against other state-of-the-art foundation models in various data sets while scaling significantly better with the prediction length. The source code can be accessed at https://github.com/automl/Mamba4Cast.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/b3f1f2c4fbda68dc648799ee98f7bd4c26489a7e" target='_blank'>
-              Simplified Mamba with Disentangled Dependency Encoding for Long-Term Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/bf7e0c212f62d082df600d79a5e7c207f5094ed9" target='_blank'>
+              Mamba4Cast: Efficient Zero-Shot Time Series Forecasting with State Space Models
               </a>
             </td>
           <td>
-            Zixuan Weng, Jindong Han, Wenzhao Jiang, Hao Liu
+            Sathya Kamesh Bhethanabhotla, Omar Swelam, Julien N. Siems, David Salinas, Frank Hutter
           </td>
-          <td>2024-08-22</td>
+          <td>2024-10-12</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>10</td>
+          <td>8</td>
         </tr>
 
-        <tr id="Long time series forecasting aims to utilize historical information to forecast future states over extended horizons. Traditional RNN-based series forecasting methods struggle to effectively address long-term dependencies and gradient issues in long time series problems. Recently, SegRNN has emerged as a leading RNN-based model tailored for long-term series forecasting, demonstrating state-of-the-art performance while maintaining a streamlined architecture through innovative segmentation and parallel decoding techniques. Nevertheless, SegRNN has several limitations: its fixed segmentation disrupts data continuity and fails to effectively leverage information across different segments, the segmentation strategy employed by SegRNN does not fundamentally address the issue of information loss within the recurrent structure. To address these issues, we propose the ISMRNN method with three key enhancements: we introduce an implicit segmentation structure to decompose the time series and map it to segmented hidden states, resulting in denser information exchange during the segmentation phase. Additionally, we incorporate residual structures in the encoding layer to mitigate information loss within the recurrent structure. To extract information more effectively, we further integrate the Mamba architecture to enhance time series information extraction. Experiments on several real-world long time series forecasting datasets demonstrate that our model surpasses the performance of current state-of-the-art models.">
+        <tr id="We present Timer-XL, a generative Transformer for unified time series forecasting. To uniformly predict 1D and 2D time series, we generalize next token prediction, predominantly adopted for causal generation of 1D sequences, to multivariate next token prediction. The proposed paradigm uniformly formulates various forecasting scenarios as a long-context generation problem. We opt for the generative Transformer, which can capture global-range and causal dependencies while providing contextual flexibility, to implement unified forecasting on univariate series characterized by non-stationarity, multivariate time series with complicated dynamics and correlations, and covariate-informed contexts that include both endogenous and exogenous variables. Technically, we propose a universal TimeAttention to facilitate generative Transformers on time series, which can effectively capture fine-grained intra- and inter-series dependencies of flattened time series tokens (patches) and is further strengthened by position embeddings in both temporal and variable dimensions. Timer-XL achieves state-of-the-art performance across challenging forecasting benchmarks through a unified approach. As a large time series model, it demonstrates notable model transferability by large-scale pre-training, as well as contextual flexibility in token lengths, positioning it as a one-for-all forecaster.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/9f37604f928c48d5acef8b85dbddb518050f7da9" target='_blank'>
-              ISMRNN: An Implicitly Segmented RNN Method with Mamba for Long-Term Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/518ecc43f230938e62b1f237ce9a0ef0d1f6d449" target='_blank'>
+              Timer-XL: Long-Context Transformers for Unified Time Series Forecasting
               </a>
             </td>
           <td>
-            Gaoxiang Zhao, Li Zhou, Xiaoqiang Wang
+            Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long
           </td>
-          <td>2024-07-15</td>
+          <td>2024-10-07</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>67</td>
         </tr>
 
-        <tr id="Recent studies have raised questions about the suitability of the Transformer architecture for long sequence time-series forecasting. These forecasting models leverage Transformers to capture dependencies between multiple time steps in a time series, with embedding tokens composed of data from individual time steps. However, challenges arise when applying Transformers to predict long sequences with strong periodicity, leading to performance degradation and increased computational burden. Furthermore, embedding tokens formed one time step at a time may struggle to reveal meaningful information in long sequences, failing to capture correlations between different time steps. In this study, we propose Skip-Timeformer, a Transformer-based model that utilizes a skip-time interaction for long sequence time-series forecasting. Specifically, we decompose the time series into multiple subsequences based on different time intervals, embedding various time steps into variable tokens across multiple sequences. The skip-time interaction mechanism utilizes these variable tokens to capture dependencies in the skip-time dimension. Additionally, skip-time interaction is employed to learn dependencies between sequences missed by multiple skip time steps. The Skip-Timeformer model demonstrates state-of-the-art performance on various real-world datasets, further enhancing the long sequence forecasting capabilities of the Transformer variations and better adapting to arbitrary lookback windows.">
+        <tr id="Time series data is prevalent across numerous fields, necessitating the development of robust and accurate forecasting models. Capturing patterns both within and between temporal and multivariate components is crucial for reliable predictions. We introduce xLSTM-Mixer, a model designed to effectively integrate temporal sequences, joint time-variate information, and multiple perspectives for robust forecasting. Our approach begins with a linear forecast shared across variates, which is then refined by xLSTM blocks. These blocks serve as key elements for modeling the complex dynamics of challenging time series data. xLSTM-Mixer ultimately reconciles two distinct views to produce the final forecast. Our extensive evaluations demonstrate xLSTM-Mixer's superior long-term forecasting performance compared to recent state-of-the-art methods. A thorough model analysis provides further insights into its key components and confirms its robustness and effectiveness. This work contributes to the resurgence of recurrent models in time series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/62917f180f12405293773a5f3689236727699caf" target='_blank'>
-              Skip-Timeformer: Skip-Time Interaction Transformer for Long Sequence Time-Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/900a5c5ff13bd734c5142a9cee2ef5df29714296" target='_blank'>
+              xLSTM-Mixer: Multivariate Time Series Forecasting by Mixing via Scalar Memories
               </a>
             </td>
           <td>
-            Wenchang Zhang, Hua Wang, Fan Zhang
+            Maurice Kraus, Felix Divo, D. Dhami, K. Kersting
           </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
+          <td>2024-10-22</td>
+          <td>ArXiv</td>
           <td>0</td>
-          <td>4</td>
+          <td>12</td>
         </tr>
 
-        <tr id="Financial time series modeling is crucial for understanding and predicting market behaviors but faces challenges such as non-linearity, non-stationarity, and high noise levels. Traditional models struggle to capture complex patterns due to these issues, compounded by limitations in computational resources and model capacity. Inspired by the success of large language models in NLP, we introduce $\textbf{PLUTUS}$, a $\textbf{P}$re-trained $\textbf{L}$arge $\textbf{U}$nified $\textbf{T}$ransformer-based model that $\textbf{U}$nveils regularities in financial time $\textbf{S}$eries. PLUTUS uses an invertible embedding module with contrastive learning and autoencoder techniques to create an approximate one-to-one mapping between raw data and patch embeddings. TimeFormer, an attention based architecture, forms the core of PLUTUS, effectively modeling high-noise time series. We incorporate a novel attention mechanisms to capture features across both variable and temporal dimensions. PLUTUS is pre-trained on an unprecedented dataset of 100 billion observations, designed to thrive in noisy financial environments. To our knowledge, PLUTUS is the first open-source, large-scale, pre-trained financial time series model with over one billion parameters. It achieves state-of-the-art performance in various tasks, demonstrating strong transferability and establishing a robust foundational model for finance. Our research provides technical guidance for pre-training financial time series data, setting a new standard in the field.">
+        <tr id="Motivated by the recent success of time-series foundation models for zero-shot forecasting, we present a methodology for $\textit{in-context fine-tuning}$ of a time-series foundation model. In particular, we design a pretrained foundation model that can be prompted (at inference time) with multiple time-series examples, in order to forecast a target time-series into the future. Our foundation model is specifically trained to utilize examples from multiple related time-series in its context window (in addition to the history of the target time-series) to help it adapt to the specific distribution of the target domain at inference time. We show that such a foundation model that uses in-context examples at inference time can obtain much better performance on popular forecasting benchmarks compared to supervised deep learning methods, statistical models, as well as other time-series foundation models. Interestingly, our in-context fine-tuning approach even rivals the performance of a foundation model that is explicitly fine-tuned on the target domain.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4a10da828bfc127919791edb726965220ece2055" target='_blank'>
-              PLUTUS: A Well Pre-trained Large Unified Transformer can Unveil Financial Time Series Regularities
+          <td><a href="https://www.semanticscholar.org/paper/1216a29c0beb36a2b621b59d36877edf05afcf8c" target='_blank'>
+              In-Context Fine-Tuning for Time-Series Foundation Models
               </a>
             </td>
           <td>
-            Yuanjian Xu, Anxian Liu, Jianing Hao, Zhenzhuo Li, Shichang Meng, Guang Zhang
+            Abhimanyu Das, Matthew Faw, Rajat Sen, Yichen Zhou
           </td>
-          <td>2024-08-19</td>
+          <td>2024-10-31</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>14</td>
         </tr>
 
-        <tr id="Time series modeling is crucial for many applications, however, it faces challenges such as complex spatio-temporal dependencies and distribution shifts in learning from historical context to predict task-specific outcomes. To address these challenges, we propose a novel approach using an agentic Retrieval-Augmented Generation (RAG) framework for time series analysis. The framework leverages a hierarchical, multi-agent architecture where the master agent orchestrates specialized sub-agents and delegates the end-user request to the relevant sub-agent. The sub-agents utilize smaller, pre-trained language models (SLMs) customized for specific time series tasks through fine-tuning using instruction tuning and direct preference optimization, and retrieve relevant prompts from a shared repository of prompt pools containing distilled knowledge about historical patterns and trends to improve predictions on new data. Our proposed modular, multi-agent RAG approach offers flexibility and achieves state-of-the-art performance across major time series tasks by tackling complex challenges more effectively than task-specific customized methods across benchmark datasets.">
+        <tr id="Time series foundation models have shown impressive performance on a variety of tasks, across a wide range of domains, even in zero-shot settings. However, most of these models are designed to handle short univariate time series as an input. This limits their practical use, especially in domains such as healthcare with copious amounts of long and multivariate data with strong temporal and intra-variate dependencies. Our study bridges this gap by cataloging and systematically comparing various context expansion techniques from both language and time series domains, and introducing a novel compressive memory mechanism to allow encoder-only TSFMs to effectively model intra-variate dependencies. We demonstrate the benefits of our approach by imbuing MOMENT, a recent family of multi-task time series foundation models, with the multivariate context.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/206b0021c5c7a1acc85f2fdee0c419e780dbca74" target='_blank'>
-              Agentic Retrieval-Augmented Generation for Time Series Analysis
+          <td><a href="https://www.semanticscholar.org/paper/687b34af103f7687d64e106b27f24a08021a8583" target='_blank'>
+              Towards Long-Context Time Series Foundation Models
               </a>
             </td>
           <td>
-            Chidaksh Ravuru, Sagar Srinivas Sakhinana, Venkataramana Runkana
+            Nina Zukowska, Mononito Goswami, Michał Wiliński, Willa Potosnak, Artur Dubrawski
           </td>
-          <td>2024-08-18</td>
+          <td>2024-09-20</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>3</td>
+          <td>8</td>
         </tr>
 
-        <tr id="To achieve more accurate prediction results in Time Series Forecasting (TSF), it is essential to distinguish between the valuable patterns (invariant patterns) of the spatial-temporal relationship and the patterns that are prone to generate distribution shift (variant patterns), then combine them for forecasting.The existing works, such as transformer-based models and GNN-based models, focus on capturing main forecasting dependencies whether it is stable or not, and they tend to overlook patterns that carry both useful information and distribution shift. In this paper, we propose a model for better forecasting time series: Decoupled Invariant Attention Network (DIAN), which contains two modules to learn spatial and temporal relationships respectively: 1) Spatial Decoupled Invariant-Variant Learning (SDIVL) to decouple the spatial invariant and variant attention scores, and then leverage convolutional networks to effectively integrate them for subsequent layers; 2) Temporal Augmented Invariant-Variant Learning (TAIVL) to decouple temporal invariant and variant patterns and combine them for further forecasting.In this module, we also design Temporal Intervention Mechanism to create multiple intervened samples by reassembling variant patterns across time stamps to eliminate the spurious impacts of variant patterns.In addition, we propose Joint Optimization to minimize the loss function considering all invariant patterns, variant patterns and intervened patterns so that our model can gain a more stable predictive ability.Extensive experiments on five datasets have demonstrated our superior performance with higher efficiency compared with state-of-the-art methods.">
+        <tr id="Time series foundation models have demonstrated impressive performance as zero-shot forecasters. However, achieving effectively unified training on time series remains an open challenge. Existing approaches introduce some level of model specialization to account for the highly heterogeneous nature of time series data. For instance, Moirai pursues unified training by employing multiple input/output projection layers, each tailored to handle time series at a specific frequency. Similarly, TimesFM maintains a frequency embedding dictionary for this purpose. We identify two major drawbacks to this human-imposed frequency-level model specialization: (1) Frequency is not a reliable indicator of the underlying patterns in time series. For example, time series with different frequencies can display similar patterns, while those with the same frequency may exhibit varied patterns. (2) Non-stationarity is an inherent property of real-world time series, leading to varied distributions even within a short context window of a single time series. Frequency-level specialization is too coarse-grained to capture this level of diversity. To address these limitations, this paper introduces Moirai-MoE, using a single input/output projection layer while delegating the modeling of diverse time series patterns to the sparse mixture of experts (MoE) within Transformers. With these designs, Moirai-MoE reduces reliance on human-defined heuristics and enables automatic token-level specialization. Extensive experiments on 39 datasets demonstrate the superiority of Moirai-MoE over existing foundation models in both in-distribution and zero-shot scenarios. Furthermore, this study conducts comprehensive model analyses to explore the inner workings of time series MoE foundation models and provides valuable insights for future research.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/08c5216266696b3eea52e8c688f6e7f32711e5a6" target='_blank'>
-              Decoupled Invariant Attention Network for Multivariate Time-series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/7a944dc164895d923498347894c7ed780d944066" target='_blank'>
+              Moirai-MoE: Empowering Time Series Foundation Models with Sparse Mixture of Experts
               </a>
             </td>
           <td>
-            Haihua Xu, Wei Fan, Kun Yi, Pengyang Wang
+            Xu Liu, Juncheng Liu, Gerald Woo, Taha İbrahim Aksu, Yuxuan Liang, Roger Zimmermann, Chenghao Liu, Silvio Savarese, Caiming Xiong, Doyen Sahoo
           </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
+          <td>2024-10-14</td>
+          <td>ArXiv</td>
           <td>0</td>
-          <td>4</td>
+          <td>25</td>
         </tr>
 
-        <tr id="Deep models for Multivariate Time Series (MTS) forecasting have recently demonstrated significant success. Channel-dependent models capture complex dependencies that channel-independent models cannot capture. However, the number of channels in real-world applications outpaces the capabilities of existing channel-dependent models, and contrary to common expectations, some models underperform the channel-independent models in handling high-dimensional data, which raises questions about the performance of channel-dependent models. To address this, our study first investigates the reasons behind the suboptimal performance of these channel-dependent models on high-dimensional MTS data. Our analysis reveals that two primary issues lie in the introduced noise from unrelated series that increases the difficulty of capturing the crucial inter-channel dependencies, and challenges in training strategies due to high-dimensional data. To address these issues, we propose STHD, the Scalable Transformer for High-Dimensional Multivariate Time Series Forecasting. STHD has three components: a) Relation Matrix Sparsity that limits the noise introduced and alleviates the memory issue; b) ReIndex applied as a training strategy to enable a more flexible batch size setting and increase the diversity of training data; and c) Transformer that handles 2-D inputs and captures channel dependencies. These components jointly enable STHD to manage the high-dimensional MTS while maintaining computational feasibility. Furthermore, experimental results show STHD's considerable improvement on three high-dimensional datasets: Crime-Chicago, Wiki-People, and Traffic. The source code and dataset are publicly available https://github.com/xinzzzhou/ScalableTransformer4HighDimensionMTSF.git.">
+        <tr id="Time series forecasting remains a challenging task, particularly in the context of complex multiscale temporal patterns. This study presents LLM-Mixer, a framework that improves forecasting accuracy through the combination of multiscale time-series decomposition with pre-trained LLMs (Large Language Models). LLM-Mixer captures both short-term fluctuations and long-term trends by decomposing the data into multiple temporal resolutions and processing them with a frozen LLM, guided by a textual prompt specifically designed for time-series data. Extensive experiments conducted on multivariate and univariate datasets demonstrate that LLM-Mixer achieves competitive performance, outperforming recent state-of-the-art models across various forecasting horizons. This work highlights the potential of combining multiscale analysis and LLMs for effective and scalable time-series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c2e65d59302efd2605a7b44a9a92f5470b0660fb" target='_blank'>
-              Scalable Transformer for High Dimensional Multivariate Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/867d69684fa5692cfb29cb0f6967415f7a9ae7bd" target='_blank'>
+              LLM-Mixer: Multiscale Mixing in LLMs for Time Series Forecasting
               </a>
             </td>
           <td>
-            Xin Zhou, Weiqing Wang, W. Buntine, Shilin Qu, Abishek Sriramulu, Weicong Tan, Christoph Bergmeir
+            Md. Kowsher, Md. Shohanur, Islam Sobuj, Nusrat Jahan Prottasha, E. A. Alanis, O. Garibay, Niloofar Yousefi, Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Xuan Zhang, Xun Liang, Aakas Zhiyuli, Shusen Zhang, Rui Xu, Bo Wu. 2019, Huaqin Zhao, Zheng Liu, Zihao Wu, Yiwei Li, Tianze Yang, Peng Shu, Shaochen Xu, Haixing Dai, Lin Zhao, Gengchen Mai, Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, Wan Zhang
           </td>
-          <td>2024-08-08</td>
+          <td>2024-10-15</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>16</td>
+          <td>34</td>
         </tr>
 
-        <tr id="None">
+        <tr id="Self-supervised learning has become a popular and effective approach for enhancing time series forecasting, enabling models to learn universal representations from unlabeled data. However, effectively capturing both the global sequence dependence and local detail features within time series data remains challenging. To address this, we propose a novel generative self-supervised method called TimeDART, denoting Diffusion Auto-regressive Transformer for Time series forecasting. In TimeDART, we treat time series patches as basic modeling units. Specifically, we employ an self-attention based Transformer encoder to model the dependencies of inter-patches. Additionally, we introduce diffusion and denoising mechanisms to capture the detail locality features of intra-patch. Notably, we design a cross-attention-based denoising decoder that allows for adjustable optimization difficulty in the self-supervised task, facilitating more effective self-supervised pre-training. Furthermore, the entire model is optimized in an auto-regressive manner to obtain transferable representations. Extensive experiments demonstrate that TimeDART achieves state-of-the-art fine-tuning performance compared to the most advanced competitive methods in forecasting tasks. Our code is publicly available at https://github.com/Melmaphother/TimeDART.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/5ea74fe77e2ef3ce555be0407ca0ad8cb269e366" target='_blank'>
-              AutoCTS++: zero-shot joint neural architecture and hyperparameter search for correlated time series forecasting
+          <td><a href="https://www.semanticscholar.org/paper/0de3582c9b651ba2e9c02640b99da702617eaf21" target='_blank'>
+              Diffusion Auto-regressive Transformer for Effective Self-supervised Time Series Forecasting
               </a>
             </td>
           <td>
-            Xinle Wu, Xingjian Wu, Bin Yang, Lekui Zhou, Chenjuan Guo, Xiangfei Qiu, Jilin Hu, Zhenli Sheng, Christian S. Jensen
+            Daoyu Wang, Mingyue Cheng, Zhiding Liu, Qi Liu, Enhong Chen
           </td>
-          <td>2024-07-30</td>
-          <td>VLDB J.</td>
-          <td>0</td>
-          <td>27</td>
+          <td>2024-10-08</td>
+          <td>ArXiv</td>
+          <td>1</td>
+          <td>11</td>
         </tr>
 
-        <tr id="Advancing the capabilities of earthquake nowcasting, the real-time forecasting of seismic activities remains a crucial and enduring objective aimed at reducing casualties. This multifaceted challenge has recently gained attention within the deep learning domain, facilitated by the availability of extensive, long-term earthquake datasets. Despite significant advancements, existing literature on earthquake nowcasting lacks comprehensive evaluations of pre-trained foundation models and modern deep learning architectures. These architectures, such as transformers or graph neural networks, uniquely focus on different aspects of data, including spatial relationships, temporal patterns, and multi-scale dependencies. This paper addresses the mentioned gap by analyzing different architectures and introducing two innovation approaches called MultiFoundationQuake and GNNCoder. We formulate earthquake nowcasting as a time series forecasting problem for the next 14 days within 0.1-degree spatial bins in Southern California, spanning from 1986 to 2024. Earthquake time series is forecasted as a function of logarithm energy released by quakes. Our comprehensive evaluation employs several key performance metrics, notably Nash-Sutcliffe Efficiency and Mean Squared Error, over time in each spatial region. The results demonstrate that our introduced models outperform other custom architectures by effectively capturing temporal-spatial relationships inherent in seismic data. The performance of existing foundation models varies significantly based on the pre-training datasets, emphasizing the need for careful dataset selection. However, we introduce a new general approach termed MultiFoundationPattern that combines a bespoke pattern with foundation model results handled as auxiliary streams. In the earthquake case, the resultant MultiFoundationQuake model achieves the best overall performance.">
+        <tr id="In processing multiple time series, accounting for the individual features of each sequence can be challenging. To address this, modern deep learning methods for time series analysis combine a shared (global) model with local layers, specific to each time series, often implemented as learnable embeddings. Ideally, these local embeddings should encode meaningful representations of the unique dynamics of each sequence. However, when these are learned end-to-end as parameters of a forecasting model, they may end up acting as mere sequence identifiers. Shared processing blocks may then become reliant on such identifiers, limiting their transferability to new contexts. In this paper, we address this issue by investigating methods to regularize the learning of local learnable embeddings for time series processing. Specifically, we perform the first extensive empirical study on the subject and show how such regularizations consistently improve performance in widely adopted architectures. Furthermore, we show that methods preventing the co-adaptation of local and global parameters are particularly effective in this context. This hypothesis is validated by comparing several methods preventing the downstream models from relying on sequence identifiers, going as far as completely resetting the embeddings during training. The obtained results provide an important contribution to understanding the interplay between learnable local parameters and shared processing layers: a key challenge in modern time series processing models and a step toward developing effective foundation models for time series.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/46b36601f9d25edbd241efbb3f7b3e9eb212bd46" target='_blank'>
-              Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting
+          <td><a href="https://www.semanticscholar.org/paper/58186a3ffbc117d00588f46184acb7ae78c2bd52" target='_blank'>
+              On the Regularization of Learnable Embeddings for Time Series Processing
               </a>
             </td>
           <td>
-            Alireza Jafari, Geoffrey C. Fox, John B. Rundle, A. Donnellan, L. G. Ludwig
+            L. Butera, G. Felice, Andrea Cini, C. Alippi
           </td>
-          <td>2024-08-21</td>
+          <td>2024-10-18</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>31</td>
+          <td>50</td>
         </tr>
 
-        <tr id="In recent years, transformer-based models have gained prominence in multivariate long-term time series forecasting (LTSF), demonstrating significant advancements despite facing challenges such as high computational demands, difficulty in capturing temporal dynamics, and managing long-term dependencies. The emergence of LTSF-Linear, with its straightforward linear architecture, has notably outperformed transformer-based counterparts, prompting a reevaluation of the transformer’s utility in time series forecasting. In response, this paper presents an adaptation of a recent architecture, termed extended LSTM (xLSTM), for LTSF. xLSTM incorporates exponential gating and a revised memory structure with higher capacity that has good potential for LTSF. Our adopted architecture for LTSF, termed xLSTMTime, surpasses current approaches. We compare xLSTMTime’s performance against various state-of-the-art models across multiple real-world datasets, demonstrating superior forecasting capabilities. Our findings suggest that refined recurrent architectures can offer competitive alternatives to transformer-based models in LTSF tasks, potentially redefining the landscape of time series forecasting.">
+        <tr id="Time series forecasting remains a critical challenge across various domains, often complicated by high-dimensional data and long-term dependencies. This paper presents a novel transformer architecture for time series forecasting, incorporating two key innovations: parameter sharing (PS) and Spatial-Temporal Segment Attention (SegAtt). We also define the time series segment as the concatenation of sequence patches from the same positions across different variables. The proposed model, PSformer, reduces the number of training parameters through the parameter sharing mechanism, thereby improving model efficiency and scalability. The introduction of SegAtt could enhance the capability of capturing local spatio-temporal dependencies by computing attention over the segments, and improve global representation by integrating information across segments. The combination of parameter sharing and SegAtt significantly improves the forecasting performance. Extensive experiments on benchmark datasets demonstrate that PSformer outperforms popular baselines and other transformer-based approaches in terms of accuracy and scalability, establishing itself as an accurate and scalable tool for time series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/286a643cebfa8149f02ead76b0be31caa7d8735f" target='_blank'>
-              xLSTMTime : Long-term Time Series Forecasting With xLSTM
+          <td><a href="https://www.semanticscholar.org/paper/496f3b7be63dc5765051593b2fe1ffd19782e3ea" target='_blank'>
+              PSformer: Parameter-efficient Transformer with Segment Attention for Time Series Forecasting
               </a>
             </td>
           <td>
-            Musleh Alharthi, A. Mahmood
+            Yanlong Wang, Jian Xu, Fei Ma, Shao-Lun Huang, Danny D. Sun, Xiao-Ping Zhang
           </td>
-          <td>2024-07-14</td>
+          <td>2024-11-03</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>22</td>
+          <td>1</td>
         </tr>
 
-        <tr id="In recent years, numerous Transformer-based models have been applied to long-term time-series forecasting (LTSF) tasks. However, recent studies with linear models have questioned their effectiveness, demonstrating that simple linear layers can outperform sophisticated Transformer-based models. In this work, we review and categorize existing Transformer-based models into two main types: (1) modifications to the model structure and (2) modifications to the input data. The former offers scalability but falls short in capturing inter-sequential information, while the latter preprocesses time-series data but is challenging to use as a scalable module. We propose $\textbf{sTransformer}$, which introduces the Sequence and Temporal Convolutional Network (STCN) to fully capture both sequential and temporal information. Additionally, we introduce a Sequence-guided Mask Attention mechanism to capture global feature information. Our approach ensures the capture of inter-sequential information while maintaining module scalability. We compare our model with linear models and existing forecasting models on long-term time-series forecasting, achieving new state-of-the-art results. We also conducted experiments on other time-series tasks, achieving strong performance. These demonstrate that Transformer-based structures remain effective and our model can serve as a viable baseline for time-series tasks.">
+        <tr id="Large Language Models have demonstrated impressive performance in many pivotal web applications such as sensor data analysis. However, since LLMs are not designed for time series tasks, simpler models like linear regressions can often achieve comparable performance with far less complexity. In this study, we perform extensive experiments to assess the effectiveness of applying LLMs to key time series tasks, including forecasting, classification, imputation, and anomaly detection. We compare the performance of LLMs against simpler baseline models, such as single-layer linear models and randomly initialized LLMs. Our results reveal that LLMs offer minimal advantages for these core time series tasks and may even distort the temporal structure of the data. In contrast, simpler models consistently outperform LLMs while requiring far fewer parameters. Furthermore, we analyze existing reprogramming techniques and show, through data manifold analysis, that these methods fail to effectively align time series data with language and display pseudo-alignment behaviour in embedding space. Our findings suggest that the performance of LLM-based methods in time series tasks arises from the intrinsic characteristics and structure of time series data, rather than any meaningful alignment with the language model architecture.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/db7a68ab6a06cdccd662afaf298e8dc2b44a9b8b" target='_blank'>
-              sTransformer: A Modular Approach for Extracting Inter-Sequential and Temporal Information for Time-Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/5d1758cc07c6914d280b5690b91d04784c15d7d3" target='_blank'>
+              Revisited Large Language Model for Time Series Analysis through Modality Alignment
               </a>
             </td>
           <td>
-            Jiaheng Yin, Zhengxin Shi, Jianshen Zhang, Xiaoming Lin, Yulin Huang, Yongzhi Qi, Wei Qi
+            L. Zheng, C. Dong, W. Zhang, Lin Yue, Miao Xu, Olaf Maennel, Weitong Chen
           </td>
-          <td>2024-08-19</td>
+          <td>2024-10-16</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>2</td>
+          <td>7</td>
         </tr>
 
-        <tr id="Pre-trained Language Models (PLMs), such as ChatGPT, have significantly advanced the field of natural language processing. This progress has inspired a series of innovative studies that explore the adaptation of PLMs to time series analysis, intending to create a unified foundation model that addresses various time series analytical tasks. However, these efforts predominantly focus on Regularly Sampled Time Series (RSTS), neglecting the unique challenges posed by Irregularly Sampled Time Series (ISTS), which are characterized by non-uniform sampling intervals and prevalent missing data. To bridge this gap, this work explores the potential of PLMs for ISTS analysis. We begin by investigating the effect of various methods for representing ISTS, aiming to maximize the efficacy of PLMs in this under-explored area. Furthermore, we present a unified PLM-based framework, ISTS-PLM, which integrates time-aware and variable-aware PLMs tailored for comprehensive intra and inter-time series modeling and includes a learnable input embedding layer and a task-specific output layer to tackle diverse ISTS analytical tasks. Extensive experiments on a comprehensive benchmark demonstrate that the ISTS-PLM, utilizing a simple yet effective series-based representation for ISTS, consistently achieves state-of-the-art performance across various analytical tasks, such as classification, interpolation, and extrapolation, as well as few-shot and zero-shot learning scenarios, spanning scientific domains like healthcare and biomechanics.">
+        <tr id="Time-series forecasting has seen significant advancements with the introduction of token prediction mechanisms such as multi-head attention. However, these methods often struggle to achieve the same performance as in language modeling, primarily due to the quadratic computational cost and the complexity of capturing long-range dependencies in time-series data. State-space models (SSMs), such as Mamba, have shown promise in addressing these challenges by offering efficient solutions with linear RNNs capable of modeling long sequences with larger context windows. However, there remains room for improvement in accuracy and scalability. We propose the use of Test-Time Training (TTT) modules in a parallel architecture to enhance performance in long-term time series forecasting. Through extensive experiments on standard benchmark datasets, we demonstrate that TTT modules consistently outperform state-of-the-art models, including the Mamba-based TimeMachine, particularly in scenarios involving extended sequence and prediction lengths. Our results show significant improvements in Mean Squared Error (MSE) and Mean Absolute Error (MAE), especially on larger datasets such as Electricity, Traffic, and Weather, underscoring the effectiveness of TTT in capturing long-range dependencies. Additionally, we explore various convolutional architectures within the TTT framework, showing that even simple configurations like 1D convolution with small filters can achieve competitive results. This work sets a new benchmark for time-series forecasting and lays the groundwork for future research in scalable, high-performance forecasting models.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/7d44918747e21cdc41c2b47b1c4ea69181cd51df" target='_blank'>
-              Unleash The Power of Pre-Trained Language Models for Irregularly Sampled Time Series
+          <td><a href="https://www.semanticscholar.org/paper/c86d4019359e33d8e00633bcb81c5722924fbbbf" target='_blank'>
+              Test Time Learning for Time Series Forecasting
               </a>
             </td>
           <td>
-            Weijiao Zhang, Chenlong Yin, Hao Liu, Hui Xiong
+            Panayiotis Christou, Shichu Chen, Xupeng Chen, Parijat Dube
           </td>
-          <td>2024-08-12</td>
+          <td>2024-09-21</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>6</td>
+          <td>4</td>
         </tr>
 
-        <tr id="Long-term time series forecasting (LTSF) has been widely applied in finance, traffic prediction, and other domains. Recently, patch-based transformers have emerged as a promising approach, segmenting data into sub-level patches that serve as input tokens. However, existing methods mostly rely on predetermined patch lengths, necessitating expert knowledge and posing challenges in capturing diverse characteristics across various scales. Moreover, time series data exhibit diverse variations and fluctuations across different temporal scales, which traditional approaches struggle to model effectively. In this paper, we propose a dynamic tokenizer with a dynamic sparse learning algorithm to capture diverse receptive fields and sparse patterns of time series data. In order to build hierarchical receptive fields, we develop a multi-scale Transformer model, coupled with multi-scale sequence extraction, capable of capturing multi-resolution features. Additionally, we introduce a group-aware rotary position encoding technique to enhance intra- and inter-group position awareness among representations across different temporal scales. Our proposed model, named DRFormer, is evaluated on various real-world datasets, and experimental results demonstrate its superiority compared to existing methods. Our code is available at: https://github.com/ruixindingECNU/DRFormer.">
+        <tr id="Irregularly sampled time series forecasting, characterized by non-uniform intervals, is prevalent in practical applications. However, previous research have been focused on regular time series forecasting, typically relying on transformer architectures. To extend transformers to handle irregular time series, we tackle the positional embedding which represents the temporal information of the data. We propose CTLPE, a method learning a continuous linear function for encoding temporal information. The two challenges of irregular time series, inconsistent observation patterns and irregular time gaps, are solved by learning a continuous-time function and concise representation of position. Additionally, the linear continuous function is empirically shown superior to other continuous functions by learning a neural controlled differential equation-based positional embedding, and theoretically supported with properties of ideal positional embedding. CTLPE outperforms existing techniques across various irregularly-sampled time series datasets, showcasing its enhanced efficacy.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/93c28e3a8cfde1fbb13658d2f625053edc878a02" target='_blank'>
-              DRFormer: Multi-Scale Transformer Utilizing Diverse Receptive Fields for Long Time-Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/3def10f5aec43ee78891e66d0efb31ad4f06d5b0" target='_blank'>
+              Continuous-Time Linear Positional Embedding for Irregular Time Series Forecasting
               </a>
             </td>
           <td>
-            Ruixin Ding, Yuqi Chen, Yu-Ting Lan, Wei Zhang
+            Byunghyun Kim, Jae-Gil Lee
           </td>
-          <td>2024-08-05</td>
+          <td>2024-09-30</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>2</td>
         </tr>
 
-        <tr id="It is challenging to scale time series forecasting models such that they forecast accurately for multiple distinct domains and datasets, all with potentially different underlying collection procedures (e.g., sample resolution), patterns (e.g., periodicity), and prediction requirements (e.g., reconstruction vs. forecasting). We call this general task universal forecasting. Existing methods usually assume that input data is regularly sampled, and they forecast to pre-determined horizons, resulting in failure to generalise outside of the scope of their training. We propose the DAM - a neural model that takes randomly sampled histories and outputs an adjustable basis composition as a continuous function of time for forecasting to non-fixed horizons. It involves three key components: (1) a flexible approach for using randomly sampled histories from a long-tail distribution, that enables an efficient global perspective of the underlying temporal dynamics while retaining focus on the recent history; (2) a transformer backbone that is trained on these actively sampled histories to produce, as representational output, (3) the basis coefficients of a continuous function of time. We show that a single univariate DAM, trained on 25 time series datasets, either outperformed or closely matched existing SoTA models at multivariate long-term forecasting across 18 datasets, including 8 held-out for zero-shot transfer, even though these models were trained to specialise for each dataset-horizon combination. This single DAM excels at zero-shot transfer and very-long-term forecasting, performs well at imputation, is interpretable via basis function composition and attention, can be tuned for different inference-cost requirements, is robust to missing and irregularly sampled data {by design}.">
+        <tr id="Attention-based architectures have become ubiquitous in time series forecasting tasks, including spatio-temporal (STF) and long-term time series forecasting (LTSF). Yet, our understanding of the reasons for their effectiveness remains limited. This work proposes a new way to understand self-attention networks: we have shown empirically that the entire attention mechanism in the encoder can be reduced to an MLP formed by feedforward, skip-connection, and layer normalization operations for temporal and/or spatial modeling in multivariate time series forecasting. Specifically, the Q, K, and V projection, the attention score calculation, the dot-product between the attention score and the V, and the final projection can be removed from the attention-based networks without significantly degrading the performance that the given network remains the top-tier compared to other SOTA methods. For spatio-temporal networks, the MLP-replace-attention network achieves a reduction in FLOPS of $62.579\%$ with a loss in performance less than $2.5\%$; for LTSF, a reduction in FLOPs of $42.233\%$ with a loss in performance less than $2\%$.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/e177ce13d5e3ca4673b2fff9481408cd98bba0d3" target='_blank'>
-              DAM: Towards A Foundation Model for Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/eef80304bcbf017f066c662d896014bafb4b6cbb" target='_blank'>
+              Approximate attention with MLP: a pruning strategy for attention-based model in multivariate time series forecasting
               </a>
             </td>
           <td>
-            Luke Darlow, Qiwen Deng, Ahmed Hassan, Martin Asenov, Rajkarn Singh, Artjom Joosen, Adam Barker, A. Storkey
+            Suhan Guo, Jiahong Deng, Yi Wei, Hui Dou, F. Shen, Jian Zhao
           </td>
-          <td>2024-07-25</td>
+          <td>2024-10-31</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>43</td>
+          <td>7</td>
         </tr>
 
-        <tr id="Time series forecasting requires balancing short-term and long-term dependencies for accurate predictions. Existing methods mainly focus on long-term dependency modeling, neglecting the complexities of short-term dynamics, which may hinder performance. Transformers are superior in modeling long-term dependencies but are criticized for their quadratic computational cost. Mamba provides a near-linear alternative but is reported less effective in time series longterm forecasting due to potential information loss. Current architectures fall short in offering both high efficiency and strong performance for long-term dependency modeling. To address these challenges, we introduce Mixture of Universals (MoU), a versatile model to capture both short-term and long-term dependencies for enhancing performance in time series forecasting. MoU is composed of two novel designs: Mixture of Feature Extractors (MoF), an adaptive method designed to improve time series patch representations for short-term dependency, and Mixture of Architectures (MoA), which hierarchically integrates Mamba, FeedForward, Convolution, and Self-Attention architectures in a specialized order to model long-term dependency from a hybrid perspective. The proposed approach achieves state-of-the-art performance while maintaining relatively low computational costs. Extensive experiments on seven real-world datasets demonstrate the superiority of MoU. Code is available at https://github.com/lunaaa95/mou/.">
+        <tr id="Time series analysis plays a critical role in numerous applications, supporting tasks such as forecasting, classification, anomaly detection, and imputation. In this work, we present the time series pattern machine (TSPM), a model designed to excel in a broad range of time series tasks through powerful representation and pattern extraction capabilities. Traditional time series models often struggle to capture universal patterns, limiting their effectiveness across diverse tasks. To address this, we define multiple scales in the time domain and various resolutions in the frequency domain, employing various mixing strategies to extract intricate, task-adaptive time series patterns. Specifically, we introduce a general-purpose TSPM that processes multi-scale time series using (1) multi-resolution time imaging (MRTI), (2) time image decomposition (TID), (3) multi-scale mixing (MCM), and (4) multi-resolution mixing (MRM) to extract comprehensive temporal patterns. MRTI transforms multi-scale time series into multi-resolution time images, capturing patterns across both temporal and frequency domains. TID leverages dual-axis attention to extract seasonal and trend patterns, while MCM hierarchically aggregates these patterns across scales. MRM adaptively integrates all representations across resolutions. This method achieves state-of-the-art performance across 8 time series analytical tasks, consistently surpassing both general-purpose and task-specific models. Our work marks a promising step toward the next generation of TSPMs, paving the way for further advancements in time series analysis.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/571baa10e035fd80b918d2352932b9fdc3f27ba9" target='_blank'>
-              Mamba or Transformer for Time Series Forecasting? Mixture of Universals (MoU) Is All You Need
+          <td><a href="https://www.semanticscholar.org/paper/61937e2748bc587b2b0b7f0f7430cc0fce9bc299" target='_blank'>
+              TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis
               </a>
             </td>
           <td>
-            Sijia Peng, Yun Xiong, Yangyong Zhu, Zhiqiang Shen
+            Shiyu Wang, Jiawei Li, X. Shi, Zhou Ye, Baichuan Mo, Wenze Lin, Shengtong Ju, Zhixuan Chu, Ming Jin
           </td>
-          <td>2024-08-28</td>
+          <td>2024-10-21</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>19</td>
+          <td>3</td>
         </tr>
 
-        <tr id="In multivariate time-series forecasting (MTSF), extracting the temporal correlations of the input sequences is crucial. While popular Transformer-based predictive models can perform well, their quadratic computational complexity results in inefficiency and high overhead. The recently emerged Mamba, a selective state space model, has shown promising results in many fields due to its strong temporal feature extraction capabilities and linear computational complexity. However, due to the unilateral nature of Mamba, channel-independent predictive models based on Mamba cannot attend to the relationships among all variables in the manner of Transformer-based models. To address this issue, we combine fast-attention with Mamba to introduce a novel framework named FMamba for MTSF. Technically, we first extract the temporal features of the input variables through an embedding layer, then compute the dependencies among input variables via the fast-attention module. Subsequently, we use Mamba to selectively deal with the input features and further extract the temporal dependencies of the variables through the multi-layer perceptron block (MLP-block). Finally, FMamba obtains the predictive results through the projector, a linear layer. Experimental results on eight public datasets demonstrate that FMamba can achieve state-of-the-art performance while maintaining low computational overhead.">
+        <tr id="In natural language processing and computer vision, self-supervised pre-training on large datasets unlocks foundational model capabilities across domains and tasks. However, this potential has not yet been realised in time series analysis, where existing methods disregard the heterogeneous nature of time series characteristics. Time series are prevalent in many domains, including medicine, engineering, natural sciences, and finance, but their characteristics vary significantly in terms of variate count, inter-variate relationships, temporal dynamics, and sampling frequency. This inherent heterogeneity across domains prevents effective pre-training on large time series corpora. To address this issue, we introduce OTiS, an open model for general time series analysis, that has been specifically designed to handle multi-domain heterogeneity. We propose a novel pre-training paradigm including a tokeniser with learnable domain-specific signatures, a dual masking strategy to capture temporal causality, and a normalised cross-correlation loss to model long-range dependencies. Our model is pre-trained on a large corpus of 640,187 samples and 11 billion time points spanning 8 distinct domains, enabling it to analyse time series from any (unseen) domain. In comprehensive experiments across 15 diverse applications - including classification, regression, and forecasting - OTiS showcases its ability to accurately capture domain-specific data characteristics and demonstrates its competitiveness against state-of-the-art baselines. Our code and pre-trained weights are publicly available at https://github.com/oetu/otis.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/3c5ced872aa5d16bf3aa59c40931ed85b54cb3e5" target='_blank'>
-              FMamba: Mamba based on Fast-attention for Multivariate Time-series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/67381675766ffe648e1346cb19d0d7318366bbbe" target='_blank'>
+              Towards Generalisable Time Series Understanding Across Domains
               </a>
             </td>
           <td>
-            Shusen Ma, Yu Kang, Peng Bai, Yunan Zhao
+            Ozgun Turgut, Philip Muller, M. Menten, Daniel Rueckert
           </td>
-          <td>2024-07-20</td>
+          <td>2024-10-09</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>15</td>
         </tr>
 
-        <tr id="Discrete-Time Dynamic Graphs (DTDGs), which are prevalent in real-world implementations and notable for their ease of data acquisition, have garnered considerable attention from both academic researchers and industry practitioners. The representation learning of DTDGs has been extensively applied to model the dynamics of temporally changing entities and their evolving connections. Currently, DTDG representation learning predominantly relies on GNN+RNN architectures, which manifest the inherent limitations of both Graph Neural Networks (GNNs) and Recurrent Neural Networks (RNNs). GNNs suffer from the over-smoothing issue as the models architecture goes deeper, while RNNs struggle to capture long-term dependencies effectively. GNN+RNN architectures also grapple with scaling to large graph sizes and long sequences. Additionally, these methods often compute node representations separately and focus solely on individual node characteristics, thereby overlooking the behavior intersections between the two nodes whose link is being predicted, such as instances where the two nodes appear together in the same context or share common neighbors. This paper introduces a novel representation learning method DTFormer for DTDGs, pivoting from the traditional GNN+RNN framework to a Transformer-based architecture. Our approach exploits the attention mechanism to concurrently process topological information within the graph at each timestamp and temporal dynamics of graphs along the timestamps, circumventing the aforementioned fundamental weakness of both GNNs and RNNs. Moreover, we enhance the model's expressive capability by incorporating the intersection relationships among nodes and integrating a multi-patching module. Extensive experiments conducted on six public dynamic graph benchmark datasets confirm our model's efficacy, achieving the SOTA performance.">
+        <tr id="Recently, time series foundation models have shown promising zero-shot forecasting performance on time series from a wide range of domains. However, it remains unclear whether their success stems from a true understanding of temporal dynamics or simply from memorizing the training data. While implicit reasoning in language models has been studied, similar evaluations for time series models have been largely unexplored. This work takes an initial step toward assessing the reasoning abilities of deep time series forecasting models. We find that certain linear, MLP-based, and patch-based Transformer models generalize effectively in systematically orchestrated out-of-distribution scenarios, suggesting underexplored reasoning capabilities beyond simple pattern memorization.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/d0d562378884600e23eb9392361a2aa22e512b2e" target='_blank'>
-              DTFormer: A Transformer-Based Method for Discrete-Time Dynamic Graph Representation Learning
+          <td><a href="https://www.semanticscholar.org/paper/0fb2694a6193c3c333e8501154e0b9bd23dbb2ee" target='_blank'>
+              Implicit Reasoning in Deep Time Series Forecasting
               </a>
             </td>
           <td>
-            Xi Chen, Yun Xiong, Siwei Zhang, Jiawei Zhang, Yao Zhang, Shiyang Zhou, Xixi Wu, Mingyang Zhang, Tengfei Liu, Weiqiang Wang
+            Willa Potosnak, Cristian Challu, Mononito Goswami, Michał Wiliński, Nina Zukowska
           </td>
-          <td>2024-07-26</td>
+          <td>2024-09-17</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>4</td>
+          <td>8</td>
         </tr>
 
-        <tr id="Traditional recurrent neural network architectures, such as long short-term memory neural networks (LSTM), have historically held a prominent role in time series forecasting (TSF) tasks. While the recently introduced sLSTM for Natural Language Processing (NLP) introduces exponential gating and memory mixing that are beneficial for long term sequential learning, its potential short memory issue is a barrier to applying sLSTM directly in TSF. To address this, we propose a simple yet efficient algorithm named P-sLSTM, which is built upon sLSTM by incorporating patching and channel independence. These modifications substantially enhance sLSTM's performance in TSF, achieving state-of-the-art results. Furthermore, we provide theoretical justifications for our design, and conduct extensive comparative and analytical experiments to fully validate the efficiency and superior performance of our model.">
+        <tr id="Time series forecasting, which aims to predict future values based on historical data, has garnered significant attention due to its broad range of applications. However, real-world time series often exhibit complex non-uniform distribution with varying patterns across segments, such as season, operating condition, or semantic meaning, making accurate forecasting challenging. Existing approaches, which typically train a single model to capture all these diverse patterns, often struggle with the pattern drifts between patches and may lead to poor generalization. To address these challenges, we propose \textbf{TFPS}, a novel architecture that leverages pattern-specific experts for more accurate and adaptable time series forecasting. TFPS employs a dual-domain encoder to capture both time-domain and frequency-domain features, enabling a more comprehensive understanding of temporal dynamics. It then uses subspace clustering to dynamically identify distinct patterns across data patches. Finally, pattern-specific experts model these unique patterns, delivering tailored predictions for each patch. By explicitly learning and adapting to evolving patterns, TFPS achieves significantly improved forecasting accuracy. Extensive experiments on real-world datasets demonstrate that TFPS outperforms state-of-the-art methods, particularly in long-term forecasting, through its dynamic and pattern-aware learning approach. The data and codes are available: \url{https://github.com/syrGitHub/TFPS}.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/db9accd4e74a2c48887d572350b75ba663c679f1" target='_blank'>
-              Unlocking the Power of LSTM for Long Term Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/b535dd5cfb6225e935a51fb38d812dc907008a4c" target='_blank'>
+              Learning Pattern-Specific Experts for Time Series Forecasting Under Patch-level Distribution Shift
               </a>
             </td>
           <td>
-            Yaxuan Kong, Zepu Wang, Yuqi Nie, Tian Zhou, Stefan Zohren, Yuxuan Liang, Peng Sun, Qingsong Wen
+            Yanru Sun, Zongxia Xie, Emadeldeen Eldele, Dongyue Chen, Qinghua Hu, Min Wu
           </td>
-          <td>2024-08-19</td>
+          <td>2024-10-13</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>5</td>
+          <td>18</td>
         </tr>
 
-        <tr id="Accurate workload forecasting is critical for efficient resource management in cloud computing systems, enabling effective scheduling and autoscaling. Despite recent advances with transformer-based forecasting models, challenges remain due to the non-stationary, nonlinear characteristics of workload time series and the long-term dependencies. In particular, inconsistent performance between long-term history and near-term forecasts hinders long-range predictions. This paper proposes a novel framework leveraging self-supervised multiscale representation learning to capture both long-term and near-term workload patterns. The long-term history is encoded through multiscale representations while the near-term observations are modeled via temporal flow fusion. These representations of different scales are fused using an attention mechanism and characterized with normalizing flows to handle non-Gaussian/non-linear distributions of time series. Extensive experiments on 9 benchmarks demonstrate superiority over existing methods.">
+        <tr id="While numerous forecasters have been proposed using different network architectures, the Transformer-based models have state-of-the-art performance in time series forecasting. However, forecasters based on Transformers are still suffering from vulnerability to high-frequency signals, efficiency in computation, and bottleneck in full-spectrum utilization, which essentially are the cornerstones for accurately predicting time series with thousands of points. In this paper, we explore a novel perspective of enlightening signal processing for deep time series forecasting. Inspired by the filtering process, we introduce one simple yet effective network, namely FilterNet, built upon our proposed learnable frequency filters to extract key informative temporal patterns by selectively passing or attenuating certain components of time series signals. Concretely, we propose two kinds of learnable filters in the FilterNet: (i) Plain shaping filter, that adopts a universal frequency kernel for signal filtering and temporal modeling; (ii) Contextual shaping filter, that utilizes filtered frequencies examined in terms of its compatibility with input signals for dependency learning. Equipped with the two filters, FilterNet can approximately surrogate the linear and attention mappings widely adopted in time series literature, while enjoying superb abilities in handling high-frequency noises and utilizing the whole frequency spectrum that is beneficial for forecasting. Finally, we conduct extensive experiments on eight time series forecasting benchmarks, and experimental results have demonstrated our superior performance in terms of both effectiveness and efficiency compared with state-of-the-art methods. Code is available at this repository: https://github.com/aikunyi/FilterNet">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/690e772c64d7f67b330d55aa96ffa09ecc7d2ae2" target='_blank'>
-              Multiscale Representation Enhanced Temporal Flow Fusion Model for Long-Term Workload Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/3af83e96cc97a134738674d3167661ae710e5eb0" target='_blank'>
+              FilterNet: Harnessing Frequency Filters for Time Series Forecasting
               </a>
             </td>
           <td>
-            Shiyu Wang, Zhixuan Chu, Yinbo Sun, Yu Liu, Yuliang Guo, Yang Chen, Huiyang Jian, Lintao Ma, Xingyu Lu, Jun Zhou
+            Kun Yi, Jingru Fei, Qi Zhang, Hui He, Shufeng Hao, Defu Lian, Wei Fan
           </td>
-          <td>2024-07-29</td>
+          <td>2024-11-03</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>4</td>
+          <td>6</td>
         </tr>
 
-        <tr id="Exploiting self-supervised learning (SSL) to extract the universal representations of time series could not only capture the natural properties of time series but also offer huge help to the downstream tasks. Nevertheless, existing time series representation learning (TSRL) methods face challenges in attaining universality. Indeed, existing methods relying solely on one SSL strategy (either contrastive learning (CL) or generative) often fall short in capturing rich semantic information for various downstream tasks. Moreover, time series exhibit diverse distributions and inherent characteristics, particularly with the common occurrence of missing values, posing a notable challenge for existing backbones in effectively handling such diverse time series data. To bridge these gaps, we propose CTRL, a framework for universal TSRL. For the first time, we employ Neural Controlled Differential Equation (NCDE) as the backbone for TSRL, which captures the continuous processes and exhibits robustness to missing data. Additionally, a dual-task SSL strategy, integrating both reconstruction and contrasting tasks, is proposed to enrich the semantic information of the learned representations. Furthermore, novel hard negative construction and false negative elimination mechanisms are proposed to improve sampling efficiency and reduce sampling bias in CL. Finally, extensive experiments demonstrate the superiority of CTRL in forecasting, classification, and imputation tasks, particularly its outstanding robustness to missing data.">
+        <tr id="Spatio-temporal forecasting is a critical component of various smart city applications, such as transportation optimization, energy management, and socio-economic analysis. Recently, several automated spatio-temporal forecasting methods have been proposed to automatically search the optimal neural network architecture for capturing complex spatio-temporal dependencies. However, the existing automated approaches suffer from expensive neural architecture search overhead, which hinders their practical use and the further exploration of diverse spatio-temporal operators in a finer granularity. In this paper, we propose AutoSTF, a decoupled automatic neural architecture search framework for cost-effective automated spatio-temporal forecasting. From the efficiency perspective, we first decouple the mixed search space into temporal space and spatial space and respectively devise representation compression and parameter-sharing schemes to mitigate the parameter explosion. The decoupled spatio-temporal search not only expedites the model optimization process but also leaves new room for more effective spatio-temporal dependency modeling. From the effectiveness perspective, we propose a multi-patch transfer module to jointly capture multi-granularity temporal dependencies and extend the spatial search space to enable finer-grained layer-wise spatial dependency search. Extensive experiments on eight datasets demonstrate the superiority of AutoSTF in terms of both accuracy and efficiency. Specifically, our proposed method achieves up to 13.48x speed-up compared to state-of-the-art automatic spatio-temporal forecasting methods while maintaining the best forecasting accuracy.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/e7373192e3c3277dc2d058d9ba25c19c722df2fb" target='_blank'>
-              An NCDE-based Framework for Universal Representation Learning of Time Series
+          <td><a href="https://www.semanticscholar.org/paper/8c7796979329a74920ae324ef6374a0c6738d726" target='_blank'>
+              AutoSTF: Decoupled Neural Architecture Search for Cost-Effective Automated Spatio-Temporal Forecasting
               </a>
             </td>
           <td>
-            Zihang Liu, Bowen Du, Junchen Ye, Xianqing Wen, Leilei Sun
+            Tengfei Lyu, Weijiao Zhang, Jinliang Deng, Hao Liu
           </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
+          <td>2024-09-25</td>
+          <td>ArXiv</td>
           <td>0</td>
-          <td>26</td>
+          <td>7</td>
         </tr>
 
-        <tr id="Frequency domain representation of time series feature offers a concise representation for handling real-world time series data with inherent complexity and dynamic nature. However, current frequency-based methods with complex operations still fall short of state-of-the-art time domain methods for general time series analysis. In this work, we present Omni-Dimensional Frequency Learner (ODFL) model based on a in depth analysis among all the three aspects of the spectrum feature: channel redundancy property among the frequency dimension, the sparse and un-salient frequency energy distribution among the frequency dimension, and the semantic diversity among the variable dimension. Technically, our method is composed of a semantic-adaptive global filter with attention to the un-salient frequency bands and partial operation among the channel dimension. Empirical results show that ODFL achieves consistent state-of-the-art in five mainstream time series analysis tasks, including short- and long-term forecasting, imputation, classification, and anomaly detection, offering a promising foundation for time series analysis.">
+        <tr id="Numerous industrial sectors necessitate models capable of providing robust forecasts across various horizons. Despite the recent strides in crafting specific architectures for time-series forecasting and developing pre-trained universal models, a comprehensive examination of their capability in accommodating varied-horizon forecasting during inference is still lacking. This paper bridges this gap through the design and evaluation of the Elastic Time-Series Transformer (ElasTST). The ElasTST model incorporates a non-autoregressive design with placeholders and structured self-attention masks, warranting future outputs that are invariant to adjustments in inference horizons. A tunable version of rotary position embedding is also integrated into ElasTST to capture time-series-specific periods and enhance adaptability to different horizons. Additionally, ElasTST employs a multi-scale patch design, effectively integrating both fine-grained and coarse-grained information. During the training phase, ElasTST uses a horizon reweighting strategy that approximates the effect of random sampling across multiple horizons with a single fixed horizon setting. Through comprehensive experiments and comparisons with state-of-the-art time-series architectures and contemporary foundation models, we demonstrate the efficacy of ElasTST's unique design elements. Our findings position ElasTST as a robust solution for the practical necessity of varied-horizon forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/de75ddcb9fb13372a31ee08af8dd80b5913d70f3" target='_blank'>
-              Omni-Dimensional Frequency Learner for General Time Series Analysis
+          <td><a href="https://www.semanticscholar.org/paper/1655d8b0d53bfdc28d81fbb6aa765118ac291460" target='_blank'>
+              ElasTST: Towards Robust Varied-Horizon Forecasting with Elastic Time-Series Transformer
               </a>
             </td>
           <td>
-            Xianing Chen.Hanting Chen, Hanting Chen, Hailin Hu
+            Jiawen Zhang, Shun Zheng, Xumeng Wen, Xiaofang Zhou, Jiang Bian, Jia Li
           </td>
-          <td>2024-07-15</td>
+          <td>2024-11-04</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>8</td>
         </tr>
 
-        <tr id="This research examines the use of Large Language Models (LLMs) in predicting time series, with a specific focus on the LLMTIME model. Despite the established effectiveness of LLMs in tasks such as text generation, language translation, and sentiment analysis, this study highlights the key challenges that large language models encounter in the context of time series prediction. We assess the performance of LLMTIME across multiple datasets and introduce classical almost periodic functions as time series to gauge its effectiveness. The empirical results indicate that while large language models can perform well in zero-shot forecasting for certain datasets, their predictive accuracy diminishes notably when confronted with diverse time series data and traditional signals. The primary finding of this study is that the predictive capacity of LLMTIME, similar to other LLMs, significantly deteriorates when dealing with time series data that contain both periodic and trend components, as well as when the signal comprises complex frequency components.">
+        <tr id="Multivariate time-series data in fields like healthcare and industry are informative but challenging due to high dimensionality and lack of labels. Recent self-supervised learning methods excel in learning rich representations without labels but struggle with disentangled embeddings and inductive bias issues like transformation-invariance. To address these challenges, we introduce TimeDRL, a framework for multivariate time-series representation learning with dual-level disentangled embeddings. TimeDRL features: (i) disentangled timestamp-level and instance-level embeddings using a [CLS] token strategy; (ii) timestamp-predictive and instance-contrastive tasks for representation learning; and (iii) avoidance of augmentation methods to eliminate inductive biases. Experiments on forecasting and classification datasets show TimeDRL outperforms existing methods, with further validation in semi-supervised settings with limited labeled data.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/11e794835a1f7d46560666d90c8eab3c24d84c05" target='_blank'>
-              An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/1f4084722f15df7440cf6161bcfc9aebe9fbf7cb" target='_blank'>
+              Self-Supervised Learning of Disentangled Representations for Multivariate Time-Series
               </a>
             </td>
           <td>
-            Rui Cao, Qiao Wang
+            Ching Chang, Chiao-Tung Chan, Wei-Yao Wang, Wenjie Peng, Tien-Fu Chen
           </td>
-          <td>2024-08-09</td>
+          <td>2024-10-16</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>3</td>
+          <td>7</td>
         </tr>
 
-        <tr id="The complexity and heterogeneity of data in many real-world applications pose significant challenges for traditional machine learning and signal processing techniques. For instance, in medicine, effective analysis of diverse physiological signals is crucial for patient monitoring and clinical decision-making and yet highly challenging. We introduce MedTsLLM, a general multimodal large language model (LLM) framework that effectively integrates time series data and rich contextual information in the form of text to analyze physiological signals, performing three tasks with clinical relevance: semantic segmentation, boundary detection, and anomaly detection in time series. These critical tasks enable deeper analysis of physiological signals and can provide actionable insights for clinicians. We utilize a reprogramming layer to align embeddings of time series patches with a pretrained LLM's embedding space and make effective use of raw time series, in conjunction with textual context. Given the multivariate nature of medical datasets, we develop methods to handle multiple covariates. We additionally tailor the text prompt to include patient-specific information. Our model outperforms state-of-the-art baselines, including deep learning models, other LLMs, and clinical methods across multiple medical domains, specifically electrocardiograms and respiratory waveforms. MedTsLLM presents a promising step towards harnessing the power of LLMs for medical time series analysis that can elevate data-driven tools for clinicians and improve patient outcomes.">
+        <tr id="Although the Transformer-based models have been the dominant architecture for long-term time series forecasting tasks in recent years, there is an essential problem. Most of these methods adopt the standard self-attention mechanism which not only has a high computational complexity, but also fails to capture the local and global dependencies from historical time series data simultaneously. In this paper, we propose CIPformer, a novel Transformer-based model for long-term time series forecasting. It utilizes channel independence, where each channel comprises a solitary univariate time series, sharing identical embedding and Transformer weights across all series. Additionally, patch embedding is employed to extract the local information and reduce time and memory complexity of the attention mechanism quadratically. With a deftly designed patch-wise attention which encompasses both intra-patch and inter-patch attention, CIPformer empowers the Transformer encoder with the ability to capture both local features and global interactions. Furthermore, in order to learn linear information, we employ a linear model to better capture linear trends and details. Extensive experiments conducted on six real-world datasets have demonstrated that our proposed model CIPformer can enhance the multivariate long-term time series forecasting performance, when compared with some of the state-of-the-art Transformer-based methods.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/6ac9e9c8bdd4eebe3cf6adf7187b9ffb16ff5b64" target='_blank'>
-              MedTsLLM: Leveraging LLMs for Multimodal Medical Time Series Analysis
+          <td><a href="https://www.semanticscholar.org/paper/82bc5a9a2af84ca0044dcf9fee4154941af7e9a1" target='_blank'>
+              CIPformer: Channel-Independence Integrated with Patching Transformer for Multivariate Long-Term Time Series Forecasting
               </a>
             </td>
           <td>
-            Nimeesha Chan, Felix Parker, William Bennett, Tianyi Wu, Mung Yao Jia, James Fackler, Kimia Ghobadi
+            Peng Qiu, Jinhong Li
           </td>
-          <td>2024-08-14</td>
-          <td>ArXiv</td>
+          <td>2024-09-20</td>
+          <td>2024 IEEE 7th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC)</td>
+          <td>0</td>
           <td>0</td>
-          <td>1</td>
         </tr>
 
-        <tr id="Short fixed-length inputs are the main bottleneck of deep learning methods in long time-series forecasting tasks. Prolonging input length causes overfitting, rapidly deteriorating accuracy. Our research indicates that the overfitting is a combination reaction of the multi-scale pattern coupling in time series and the fixed focusing scale of current models. First, we find that the patterns exhibited by a time series across various scales are reflective of its multi-periodic nature, where each scale corresponds to specific period length. Second, We find that the token size predominantly dictates model behavior, as it determines the scale at which the model focuses and the context size it can accommodate. Our idea is to decouple the multi-scale temporal patterns of time series and to model each pattern with its corresponding period length as token size. We introduced a novel series-decomposition module(MPSD), and a Multi-Token Pattern Recognition neural network(MTPR), enabling the model to handle \textit{inputs up to $10\times$ longer}. Sufficient context enhances performance(\textit{38% maximum precision improvement}), and the decoupling approach offers \textit{Low complexity($0.22\times$ cost)} and \textit{high interpretability}.">
+        <tr id="Multivariate Time series forecasting is crucial in domains such as transportation, meteorology, and finance, especially for predicting extreme weather events. State-of-the-art methods predominantly rely on Transformer architectures, which utilize attention mechanisms to capture temporal dependencies. However, these methods are hindered by quadratic time complexity, limiting the model's scalability with respect to input sequence length. This significantly restricts their practicality in the real world. Mamba, based on state space models (SSM), provides a solution with linear time complexity, increasing the potential for efficient forecasting of sequential data. In this study, we propose UmambaTSF, a novel long-term time series forecasting framework that integrates multi-scale feature extraction capabilities of U-shaped encoder-decoder multilayer perceptrons (MLP) with Mamba's long sequence representation. To improve performance and efficiency, the Mamba blocks introduced in the framework adopt a refined residual structure and adaptable design, enabling the capture of unique temporal signals and flexible channel processing. In the experiments, UmambaTSF achieves state-of-the-art performance and excellent generality on widely used benchmark datasets while maintaining linear time complexity and low memory consumption.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/380b86c3909fe3e098a4ec9618f09d59bb9ef626" target='_blank'>
-              Long Input Sequence Network for Long Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/3d264e1c87378110d654ebbd6571cbe63c78f877" target='_blank'>
+              UmambaTSF: A U-shaped Multi-Scale Long-Term Time Series Forecasting Method Using Mamba
               </a>
             </td>
           <td>
-            Chao Ma, Yikai Hou, Xiang Li, Yinggang Sun, Haining Yu
+            Li Wu, Wenbin Pei, Jiulong Jiao, Qiang Zhang
           </td>
-          <td>2024-07-18</td>
+          <td>2024-10-15</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
-        </tr>
-
-        <tr id="Transformer has gained widespread adoption in modeling time series due to the exceptional ability of its self-attention mechanism in capturing long-range dependencies. However, when processing time series data with numerous variates, the vanilla self-attention mechanism tends to distribute attention weights evenly and smoothly, causing row-homogenization in attention maps and further hampering time series forecasting. To tackle this issue, we propose an advanced Transformer architecture entitled SDformer, which designs two novel modules, Spectral-Filter-Transform (SFT) and Dynamic-Directional-Attention (DDA), and integrates them into the encoder of Transformer to achieve more intensive attention allocation. Specifically, the SFT module utilizes the Fast Fourier Transform to select the most prominent frequencies, along with a Hamming Window to smooth and denoise the filtered series data; The DDA module applies a specialized kernel function to the query and key vectors projected from the denoised data, concentrating this innovative attention mechanism more effectively on the most informative variates to obtain a sharper attention distribution. These two modules jointly enable attention weights to be more salient among numerous variates, which in turn enhances the attention's ability to capture multivariate correlations, improving the performance in forecasting. Extensive experiments on public datasets demonstrate its superior performance over other state-of-the-art models. Code is available at https://github.com/zhouziyu02/SDformer.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2059f2f21fb401de3268af0691764082ba5fdfbd" target='_blank'>
-              SDformer: Transformer with Spectral Filter and Dynamic Attention for Multivariate Time Series Long-term Forecasting
-              </a>
-            </td>
-          <td>
-            Ziyu Zhou, Gengyu Lyu, Yiming Huang, Zihao Wang, Ziyu Jia, Zhen Yang
-          </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
           <td>0</td>
-          <td>11</td>
         </tr>
 
-        <tr id="Current Transformer methods for Multivariate Time-Series Forecasting (MTSF) are all based on the conventional attention mechanism. They involve sequence embedding and performing a linear projection of Q, K, and V, and then computing attention within this latent space. We have never delved into the attention mechanism to explore whether such a mapping space is optimal for MTSF. To investigate this issue, this study first proposes Frequency Spectrum attention (FSatten), a novel attention mechanism based on the frequency domain space. It employs the Fourier transform for embedding and introduces Multi-head Spectrum Scaling (MSS) to replace the conventional linear mapping of Q and K. FSatten can accurately capture the periodic dependencies between sequences and outperform the conventional attention without changing mainstream architectures. We further design a more general method dubbed Scaled Orthogonal attention (SOatten). We propose an orthogonal embedding and a Head-Coupling Convolution (HCC) based on the neighboring similarity bias to guide the model in learning comprehensive dependency patterns. Experiments show that FSatten and SOatten surpass the SOTA which uses conventional attention, making it a good alternative as a basic attention mechanism for MTSF. The codes and log files will be released at: https://github.com/Joeland4/FSatten-SOatten.">
+        <tr id="Accurate long-term predictions are the foundations for many machine learning applications and decision-making processes. Traditional time series approaches for prediction often focus on either autoregressive modeling, which relies solely on past observations of the target ``endogenous variables'', or forward modeling, which considers only current covariate drivers ``exogenous variables''. However, effectively integrating past endogenous and past exogenous with current exogenous variables remains a significant challenge. In this paper, we propose ExoTST, a novel transformer-based framework that effectively incorporates current exogenous variables alongside past context for improved time series prediction. To integrate exogenous information efficiently, ExoTST leverages the strengths of attention mechanisms and introduces a novel cross-temporal modality fusion module. This module enables the model to jointly learn from both past and current exogenous series, treating them as distinct modalities. By considering these series separately, ExoTST provides robustness and flexibility in handling data uncertainties that arise from the inherent distribution shift between historical and current exogenous variables. Extensive experiments on real-world carbon flux datasets and time series benchmarks demonstrate ExoTST's superior performance compared to state-of-the-art baselines, with improvements of up to 10\% in prediction accuracy. Moreover, ExoTST exhibits strong robustness against missing values and noise in exogenous drivers, maintaining consistent performance in real-world situations where these imperfections are common.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/0b4ec3cddfc06c1f060d66a3123f8a9a0b157d23" target='_blank'>
-              Revisiting Attention for Multivariate Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/dd75133dfb352d5c7f1dcd5cd33db3c149c0e557" target='_blank'>
+              ExoTST: Exogenous-Aware Temporal Sequence Transformer for Time Series Prediction
               </a>
             </td>
           <td>
-            Haixiang Wu
+            Kshitij Tayal, Arvind Renganathan, X. Jia, Vipin Kumar, Dan Lu
           </td>
-          <td>2024-07-18</td>
+          <td>2024-10-16</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>19</td>
         </tr>
 
-        <tr id="Accurate traffic flow forecasting is a crucial component of intelligent transportation systems, playing a pivotal role in enhancing transportation intelligence. The integration of Graph Neural Networks (GNNs) and Transformers in traffic flow forecasting has gained significant adoption for enhancing prediction accuracy. Yet, the complex spatial and temporal dependencies present in traffic data continue to pose substantial challenges: (1) Most GNN-based methods assume that the graph structure reflects the actual dependencies between nodes, overlooking the complex dependencies present in the real-world context. (2) Standard time-series models are unable to effectively model complex temporal dependencies, hindering prediction accuracy. To tackle these challenges, the authors propose a novel Transformer-based Spatiotemporal Graph Diffusion Convolution Network (TSGDC) for Traffic Flow Forecasting, which leverages graph diffusion and transformer to capture the complexity and dynamics of spatial and temporal patterns, thereby enhancing prediction performance. The authors designed an Efficient Channel Attention (ECA) that learns separately from the feature dimensions collected by traffic sensors and the temporal dimensions of traffic data, aiding in spatiotemporal modeling. Chebyshev Graph Diffusion Convolution (GDC) is used to capture the complex dependencies within the spatial distribution. Sequence decomposition blocks, as internal operations of transformers, are employed to gradually extract long-term stable trends from hidden complex variables. Additionally, by integrating multi-scale dependencies, including recent, daily, and weekly patterns, accurate traffic flow predictions are achieved. Experimental results on various public datasets show that TSGDC outperforms conventional traffic forecasting models, particularly in accuracy and robustness.">
+        <tr id="Long-short range time series forecasting is essential for predicting future trends and patterns over extended periods. While deep learning models such as Transformers have made significant strides in advancing time series forecasting, they often encounter difficulties in capturing long-term dependencies and effectively managing sparse semantic features. The state-space model, Mamba, addresses these issues through its adept handling of selective input and parallel computing, striking a balance between computational efficiency and prediction accuracy. This article examines the advantages and disadvantages of both Mamba and Transformer models, and introduces a combined approach, MAT, which leverages the strengths of each model to capture unique long-short range dependencies and inherent evolutionary patterns in multivariate time series. Specifically, MAT harnesses the long-range dependency capabilities of Mamba and the short-range characteristics of Transformers. Experimental results on benchmark weather datasets demonstrate that MAT outperforms existing comparable methods in terms of prediction accuracy, scalability, and memory efficiency.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/f28ffc0b410cb4e7b7c33a98e7941875942be67a" target='_blank'>
-              Transformer-Based Spatiotemporal Graph Diffusion Convolution Network for Traffic Flow Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/3e0aa05e9c0ee4fc6b3c67887960a9ef18c502d4" target='_blank'>
+              Integration of Mamba and Transformer - MAT for Long-Short Range Time Series Forecasting with Application to Weather Dynamics
               </a>
             </td>
           <td>
-            Siwei Wei, Yang Yang, Donghua Liu, Ke Deng, Chunzhi Wang
+            Wenqing Zhang, Junming Huang, Ruotong Wang, Changsong Wei, Wenqian Huang, 
           </td>
-          <td>2024-08-09</td>
-          <td>Electronics</td>
+          <td>2024-09-13</td>
+          <td>ArXiv</td>
           <td>0</td>
           <td>2</td>
         </tr>
 
-        <tr id="Graph neural networks (GNN) have shown significant capabilities in handling structured data, yet their application to dynamic, temporal data remains limited. This paper presents a new type of graph attention network, called TempoKGAT, which combines time-decaying weight and a selective neighbor aggregation mechanism on the spatial domain, which helps uncover latent patterns in the graph data. In this approach, a top-k neighbor selection based on the edge weights is introduced to represent the evolving features of the graph data. We evaluated the performance of our TempoKGAT on multiple datasets from the traffic, energy, and health sectors involving spatio-temporal data. We compared the performance of our approach to several state-of-the-art methods found in the literature on several open-source datasets. Our method shows superior accuracy on all datasets. These results indicate that TempoKGAT builds on existing methodologies to optimize prediction accuracy and provide new insights into model interpretation in temporal contexts.">
+        <tr id="Scaling laws offer valuable insights into the design of time series foundation models (TSFMs). However, previous research has largely focused on the scaling laws of TSFMs for in-distribution (ID) data, leaving their out-of-distribution (OOD) scaling behavior and the influence of model architectures less explored. In this work, we examine two common TSFM architectures, encoder-only and decoder-only Transformers, and investigate their scaling behavior on both ID and OOD data. These models are trained and evaluated across varying parameter counts, compute budgets, and dataset sizes. Our experiments reveal that the log-likelihood loss of TSFMs exhibits similar scaling behavior in both OOD and ID settings. We further compare the scaling properties across different architectures, incorporating two state-of-the-art TSFMs as case studies, showing that model architecture plays a significant role in scaling. The encoder-only Transformers demonstrate better scalability than the decoder-only Transformers, while the architectural enhancements in the two advanced TSFMs primarily improve ID performance but reduce OOD scalability. While scaling up TSFMs is expected to drive performance breakthroughs, the lack of a comprehensive understanding of TSFM scaling laws has hindered the development of a robust framework to guide model scaling. We fill this gap in this work by synthesizing our findings and providing practical guidelines for designing and scaling larger TSFMs with enhanced model capabilities.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c5c6224e833a1561925768e77e0897bc75e12cce" target='_blank'>
-              TempoKGAT: A Novel Graph Attention Network Approach for Temporal Graph Analysis
+          <td><a href="https://www.semanticscholar.org/paper/a87d911bee64f961730142670dadf9f5b8cc9210" target='_blank'>
+              Towards Neural Scaling Laws for Time Series Foundation Models
               </a>
             </td>
           <td>
-            L. Sasal, Daniel Busby, Abdenour Hadid
+            Qingren Yao, Chao-Han Huck Yang, Renhe Jiang, Yuxuan Liang, Ming Jin, Shirui Pan
           </td>
-          <td>2024-08-29</td>
+          <td>2024-10-16</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>2</td>
+          <td>6</td>
         </tr>
 
-        <tr id="Graph neural networks have shown widespread success for learning on graphs, but they still face fundamental drawbacks, such as limited expressive power, over-smoothing, and over-squashing. Meanwhile, the transformer architecture offers a potential solution to these issues. However, existing graph transformers primarily cater to homogeneous graphs and are unable to model the intricate semantics of heterogeneous graphs. Moreover, unlike small molecular graphs where the entire graph can be considered as the receptive field in graph transformers, real-world heterogeneous graphs comprise a significantly larger number of nodes and cannot be entirely treated as such. Consequently, existing graph transformers struggle to capture the long-range dependencies in these complex heterogeneous graphs. To address these two limitations, we present Poly-tokenized Heterogeneous Graph Transformer (PHGT), a novel transformer-based heterogeneous graph model. In addition to traditional node tokens, PHGT introduces a novel poly-token design with two more token types: semantic tokens and global tokens. Semantic tokens encapsulate high-order heterogeneous semantic relationships, while global tokens capture semantic-aware long-range interactions. We validate the effectiveness of PHGT through extensive experiments on standardized heterogeneous graph benchmarks, demonstrating significant improvements over state-of-the-art heterogeneous graph representation learning models.">
+        <tr id="Spatiotemporal (ST) graph modeling has garnered increasing attention recently. Most existing methods rely on a predefined graph structure or construct a single learnable graph throughout training. However, it is challenging to use a predefined graph structure to capture dynamic ST changes effectively due to evolving node relationships over time. Furthermore, these methods typically utilize only the original data, neglecting external temporal factors. Therefore, we put forward a novel time-varying graph convolutional network model that integrates external factors for multifeature ST series prediction. Firstly, we construct a time-varying adjacency matrix using attention to capture dynamic spatial relationships among nodes. The graph structure adapts over time during training, validation, and testing phases. Then, we model temporal dependence by dilated causal convolution, leveraging gated activation unit and residual connection. Notably, the prediction accuracy is enhanced through the incorporation of embedding absolute time and the fusion of multifeature. This model has been applied to three real-world multifeature datasets, achieving state-of-the-art performance in all cases. Experiments show that the method has high accuracy and robustness when applied to multifeature and multivariate ST series problems.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/b8d77a0ce1a2d638606161209e14052257a2ee23" target='_blank'>
-              Heterogeneous Graph Transformer with Poly-Tokenization
+          <td><a href="https://www.semanticscholar.org/paper/8767500a1efaa55cc4c19167ae35e6f2419f2b6a" target='_blank'>
+              TVGCN: Time-varying graph convolutional networks for multivariate and multifeature spatiotemporal series prediction
               </a>
             </td>
           <td>
-            Zhiyuan Lu, Yuan Fang, Cheng Yang, Chuan Shi
+            Feiyan Sun, Wenning Hao, Ao Zou, Kai Cheng
           </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
-          <td>1</td>
-          <td>4</td>
+          <td>2024-07-01</td>
+          <td>Science Progress</td>
+          <td>0</td>
+          <td>2</td>
         </tr>
 
-        <tr id="The non-stationary nature of real-world Multivariate Time Series (MTS) data presents forecasting models with a formidable challenge of the time-variant distribution of time series, referred to as distribution shift. Existing studies on the distribution shift mostly adhere to adaptive normalization techniques for alleviating temporal mean and covariance shifts or time-variant modeling for capturing temporal shifts. Despite improving model generalization, these normalization-based methods often assume a time-invariant transition between outputs and inputs but disregard specific intra-/inter-series correlations, while time-variant models overlook the intrinsic causes of the distribution shift. This limits model expressiveness and interpretability of tackling the distribution shift for MTS forecasting. To mitigate such a dilemma, we present a unified Probabilistic Graphical Model to Jointly capturing intra-/inter-series correlations and modeling the time-variant transitional distribution, and instantiate a neural framework called JointPGM for non-stationary MTS forecasting. Specifically, JointPGM first employs multiple Fourier basis functions to learn dynamic time factors and designs two distinct learners: intra-series and inter-series learners. The intra-series learner effectively captures temporal dynamics by utilizing temporal gates, while the inter-series learner explicitly models spatial dynamics through multi-hop propagation, incorporating Gumbel-softmax sampling. These two types of series dynamics are subsequently fused into a latent variable, which is inversely employed to infer time factors, generate final prediction, and perform reconstruction. We validate the effectiveness and efficiency of JointPGM through extensive experiments on six highly non-stationary MTS datasets, achieving state-of-the-art forecasting performance of MTS forecasting.">
+        <tr id="Multivariate time series (MTS) forecasting plays a crucial role in various real-world applications, yet simultaneously capturing both temporal and inter-variable dependencies remains a challenge. Conventional Channel-Dependent (CD) models handle these dependencies separately, limiting their ability to model complex interactions such as lead-lag dynamics. To address these limitations, we propose TiVaT (Time-Variable Transformer), a novel architecture that integrates temporal and variate dependencies through its Joint-Axis (JA) attention mechanism. TiVaT's ability to capture intricate variate-temporal dependencies, including asynchronous interactions, is further enhanced by the incorporation of Distance-aware Time-Variable (DTV) Sampling, which reduces noise and improves accuracy through a learned 2D map that focuses on key interactions. TiVaT effectively models both temporal and variate dependencies, consistently delivering strong performance across diverse datasets. Notably, it excels in capturing complex patterns within multivariate time series, enabling it to surpass or remain competitive with state-of-the-art methods. This positions TiVaT as a new benchmark in MTS forecasting, particularly in handling datasets characterized by intricate and challenging dependencies.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c259b05da12062cdb1e96cd3994b71a1b4841e89" target='_blank'>
-              Robust Multivariate Time Series Forecasting against Intra- and Inter-Series Transitional Shift
+          <td><a href="https://www.semanticscholar.org/paper/8959127abeb1daeffd9ee3243b463d8b434881db" target='_blank'>
+              TiVaT: Joint-Axis Attention for Time Series Forecasting with Lead-Lag Dynamics
               </a>
             </td>
           <td>
-            Hui He, Qi Zhang, Kun Yi, Xiaojun Xue, Shoujin Wang, Liang Hu, Longbin Cao
+            Junwoo Ha, Hyukjae Kwon, Sungsoo Kim, Kisu Lee, Ha Young Kim
           </td>
-          <td>2024-07-18</td>
+          <td>2024-10-02</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>6</td>
+          <td>3</td>
         </tr>
 
-        <tr id=". Most real-world variables are multivariate time series influenced by past values and explanatory factors. Consequently, predicting these time series data using artificial intelligence is ongoing. In particular, in fields such as healthcare and finance, where reliability is crucial, having understandable explanations for predictions is essential. However, achieving a balance between high prediction accuracy and intuitive explainability has proven challenging. Although attention-based models have limitations in representing the individual influences of each variable, these models can influence the temporal dependencies in time series prediction and the magnitude of the influence of individual variables. To address this issue, this study introduced DLFormer, an attention-based architecture integrated with distributed lag embedding, to temporally embed individual variables and capture their temporal influence. Through validation against various real-world datasets, DLFormer showcased superior performance improvements compared to existing attention-based high-performance models. Furthermore, comparing the relationships between variables enhanced the reliability of explainability.">
+        <tr id="Mamba has recently emerged as a promising alternative to Transformers, offering near-linear complexity in processing sequential data. However, while channels in time series (TS) data have no specific order in general, recent studies have adopted Mamba to capture channel dependencies (CD) in TS, introducing a sequential order bias. To address this issue, we propose SOR-Mamba, a TS forecasting method that 1) incorporates a regularization strategy to minimize the discrepancy between two embedding vectors generated from data with reversed channel orders, thereby enhancing robustness to channel order, and 2) eliminates the 1D-convolution originally designed to capture local information in sequential data. Furthermore, we introduce channel correlation modeling (CCM), a pretraining task aimed at preserving correlations between channels from the data space to the latent space in order to enhance the ability to capture CD. Extensive experiments demonstrate the efficacy of the proposed method across standard and transfer learning scenarios. Code is available at https://github.com/seunghan96/SOR-Mamba.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/69d4c5c5deda79c2d9f09e538ce80b96b9c34805" target='_blank'>
-              DLFormer: Enhancing Explainability in Multivariate Time Series Forecasting using Distributed Lag Embedding
+          <td><a href="https://www.semanticscholar.org/paper/4315bb0130a324bd6212bd8c5d1f629675aed88e" target='_blank'>
+              Sequential Order-Robust Mamba for Time Series Forecasting
               </a>
             </td>
           <td>
-            Younghwi Kim, Dohee Kim, Sunghyun Sim
+            Seunghan Lee, Juri Hong, Kibok Lee, Taeyoung Park
           </td>
-          <td>2024-08-29</td>
+          <td>2024-10-30</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>4</td>
+          <td>2</td>
         </tr>
 
-        <tr id="Long-term traffic flow forecasting plays a crucial role in intelligent transportation as it allows traffic managers to adjust their decisions in advance. However, the problem is challenging due to spatio-temporal correlations and complex dynamic patterns in continuous-time stream data. Neural Differential Equations (NDEs) are among the state-of-the-art methods for learning continuous-time traffic dynamics. However, the traditional NDE models face issues in long-term traffic forecasting due to failures in capturing delayed traffic patterns, dynamic edge (location-to-location correlation) patterns, and abrupt trend patterns. To fill this gap, we propose a new NDE architecture called Multi-View Neural Differential Equations. Our model captures current states, delayed states, and trends in different state variables (views) by learning latent multiple representations within Neural Differential Equations. Extensive experiments conducted on several real-world traffic datasets demonstrate that our proposed method outperforms the state-of-the-art and achieves superior prediction accuracy for long-term forecasting and robustness with noisy or missing inputs.">
+        <tr id="Multivariate time series forecasting plays a crucial role in various real-world applications. Significant efforts have been made to integrate advanced network architectures and training strategies that enhance the capture of temporal dependencies, thereby improving forecasting accuracy. On the other hand, mainstream approaches typically utilize a single unified model with simplistic channel-mixing embedding or cross-channel attention operations to account for the critical intricate inter-channel dependencies. Moreover, some methods even trade capacity for robust prediction based on the channel-independent assumption. Nonetheless, as time series data may display distinct evolving patterns due to the unique characteristics of each channel (including multiple strong seasonalities and trend changes), the unified modeling methods could yield suboptimal results. To this end, we propose DisenTS, a tailored framework for modeling disentangled channel evolving patterns in general multivariate time series forecasting. The central idea of DisenTS is to model the potential diverse patterns within the multivariate time series data in a decoupled manner. Technically, the framework employs multiple distinct forecasting models, each tasked with uncovering a unique evolving pattern. To guide the learning process without supervision of pattern partition, we introduce a novel Forecaster Aware Gate (FAG) module that generates the routing signals adaptively according to both the forecasters' states and input series' characteristics. The forecasters' states are derived from the Linear Weight Approximation (LWA) strategy, which quantizes the complex deep neural networks into compact matrices. Additionally, the Similarity Constraint (SC) is further proposed to guide each model to specialize in an underlying pattern by minimizing the mutual information between the representations.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/713b5e17e000120673eebab412e9b9ae591db136" target='_blank'>
-              Multi-View Neural Differential Equations for Continuous-Time Stream Data in Long-Term Traffic Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/876816dcd67c1a90461daa748acd2ccb64c35a79" target='_blank'>
+              DisenTS: Disentangled Channel Evolving Pattern Modeling for Multivariate Time Series Forecasting
               </a>
             </td>
           <td>
-            Zibo Liu, Zhe Jiang, Shigang Chen
+            Zhiding Liu, Jiqian Yang, Qingyang Mao, Yuze Zhao, Mingyue Cheng, Zhi Li, Qi Liu, Enhong Chen
           </td>
-          <td>2024-08-12</td>
+          <td>2024-10-30</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>2</td>
+          <td>5</td>
         </tr>
 
-        <tr id="Evolving networks are complex data structures that emerge in a wide range of systems in science and engineering. Learning expressive representations for such networks that encode their structural connectivity and temporal evolution is essential for downstream data analytics and machine learning applications. In this study, we introduce a self-supervised method for learning representations of temporal networks and employ these representations in the dynamic link prediction task. While temporal networks are typically characterized as a sequence of interactions over the continuous time domain, our study focuses on their discrete-time versions. This enables us to balance the trade-off between computational complexity and precise modeling of the interactions. We propose a recurrent message-passing neural network architecture for modeling the information flow over time-respecting paths of temporal networks. The key feature of our method is the contrastive training objective of the model, which is a combination of three loss functions: link prediction, graph reconstruction, and contrastive predictive coding losses. The contrastive predictive coding objective is implemented using infoNCE losses at both local and global scales of the input graphs. We empirically show that the additional self-supervised losses enhance the training and improve the model's performance in the dynamic link prediction task. The proposed method is tested on Enron, COLAB, and Facebook datasets and exhibits superior results compared to existing models.">
+        <tr id="Irregular Time Series Data (IRTS) has shown increasing prevalence in real-world applications. We observed that IRTS can be divided into two specialized types: Natural Irregular Time Series (NIRTS) and Accidental Irregular Time Series (AIRTS). Various existing methods either ignore the impacts of irregular patterns or statically learn the irregular dynamics of NIRTS and AIRTS data and suffer from limited data availability due to the sparsity of IRTS. We proposed a novel transformer-based framework for general irregular time series data that treats IRTS from four views: Locality, Time, Spatio and Irregularity to motivate the data usage to the highest potential. Moreover, we design a sophisticated irregularity-gate mechanism to adaptively select task-relevant information from irregularity, which improves the generalization ability to various IRTS data. We implement extensive experiments to demonstrate the resistance of our work to three highly missing ratio datasets (88.4\%, 94.9\%, 60\% missing value) and investigate the significance of the irregularity information for both NIRTS and AIRTS by additional ablation study. We release our implementation in https://github.com/IcurasLW/MTSFormer-Irregular_Time_Series.git">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/a362711e3ebcbbd083b9d5ea81377301c6fd9b2d" target='_blank'>
-              Contrastive Representation Learning for Dynamic Link Prediction in Temporal Networks
+          <td><a href="https://www.semanticscholar.org/paper/395ddd7773e4b19dfb97558d5903354d1124c4be" target='_blank'>
+              Irregularity-Informed Time Series Analysis: Adaptive Modelling of Spatial and Temporal Dynamics
               </a>
             </td>
           <td>
-            Amirhossein Nouranizadeh, Fatemeh Tabatabaei Far, Mohammad Rahmati
+            L. Zheng, Zhengyang Li, C. Dong, W. Zhang, Lin Yue, Miao Xu, Olaf Maennel, Weitong Chen
           </td>
-          <td>2024-08-22</td>
-          <td>ArXiv</td>
+          <td>2024-10-16</td>
+          <td>Proceedings of the 33rd ACM International Conference on Information and Knowledge Management</td>
           <td>0</td>
-          <td>2</td>
+          <td>7</td>
         </tr>
 
-        <tr id="Intelligent transportation systems based on sensor signals are crucial in addressing contemporary transportation issues, accomplishing dynamic traffic management, and facilitating route planning. However, the highly dynamic and intricate nature of traffic sensor signals presents difficulties for traffic prediction, with current models for traffic prediction inadequate in meeting the requirements of both long-term and short-term prediction tasks. In this paper, we propose a novel deep-learning framework called dynamic spatial-temporal correlation learning network (DSTLNet) that jointly leverages dynamical spatial and temporal features of traffic sensor signals to further improve the accuracy of long-and short-term traffic modeling and route planning. Specifically, we leverage the temporal convolutional network to capture long-term correlations. In addition, a spatial graph convolutional network is developed to dynamically model spatial features, and long-and short-term fusion layers are used to fuse the extracted long-and short-term temporal features, respectively. Experimental results on real-world datasets show that DSTLNet is competitive with the state-of-the-art, especially for long-term traffic prediction.">
+        <tr id="In recent years, large language models (LLMs) have shown great potential in time-series analysis by capturing complex dependencies and improving predictive performance. However, existing approaches often struggle with modality alignment, leading to suboptimal results. To address these challenges, we present a novel framework, TS-TCD, which introduces a comprehensive three-tiered cross-modal knowledge distillation mechanism. Unlike prior work that focuses on isolated alignment techniques, our framework systematically integrates: 1) Dynamic Adaptive Gating for Input Encoding and Alignment}, ensuring coherent alignment between time-series tokens and QR-decomposed textual embeddings; 2) Layer-Wise Contrastive Learning}, aligning intermediate representations across modalities to reduce feature-level discrepancies; and 3) Optimal Transport-Driven Output Alignment}, which ensures consistent output predictions through fine-grained cross-modal alignment. Extensive experiments on benchmark time-series datasets demonstrate that TS-TCD achieves state-of-the-art results, outperforming traditional methods in both accuracy and robustness.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c2ed14c7ddd2366e53463d2401366cc0c6036604" target='_blank'>
-              DSTLNet: Dynamic Spatial-Temporal Correlation Learning Network for Traffic Sensor Signal Prediction
+          <td><a href="https://www.semanticscholar.org/paper/5b3f3b64cd3b543ad8317954ebd97c7570f0449a" target='_blank'>
+              TS-TCD: Triplet-Level Cross-Modal Distillation for Time-Series Forecasting Using Large Language Models
               </a>
             </td>
           <td>
-            Yuxiang Shan, Hailiang Lu, Weidong Lou
+            Pengfei Wang, Huanran Zheng, Silong Dai, Wenjing Yue, Wei Zhu, Xiaoling Wang
           </td>
-          <td>2024-08-08</td>
-          <td>Sensors and Materials</td>
-          <td>0</td>
+          <td>2024-09-23</td>
+          <td>ArXiv</td>
+          <td>1</td>
           <td>3</td>
         </tr>
 
-        <tr id="Alongside the continuous process of improving AI performance through the development of more sophisticated models, researchers have also focused their attention to the emerging concept of data-centric AI, which emphasizes the important role of data in a systematic machine learning training process. Nonetheless, the development of models has also continued apace. One result of this progress is the development of the Transformer Architecture, which possesses a high level of capability in multiple domains such as Natural Language Processing (NLP), Computer Vision (CV) and Time Series Forecasting (TSF). Its performance is, however, heavily dependent on input data preprocessing and output data evaluation, justifying a data-centric approach to future research. We argue that data-centric AI is essential for training AI models, particularly for transformer-based TSF models efficiently. However, there is a gap regarding the integration of transformer-based TSF and data-centric AI. This survey aims to pin down this gap via the extensive literature review based on the proposed taxonomy. We review the previous research works from a data-centric AI perspective and we intend to lay the foundation work for the future development of transformer-based architecture and data-centric AI.">
+        <tr id="We propose an Autoregressive (AR) Moving-average (MA) attention structure that can adapt to various linear attention mechanisms, enhancing their ability to capture long-range and local temporal patterns in time series. In this paper, we first demonstrate that, for the time series forecasting (TSF) task, the previously overlooked decoder-only autoregressive Transformer model can achieve results comparable to the best baselines when appropriate tokenization and training methods are applied. Moreover, inspired by the ARMA model from statistics and recent advances in linear attention, we introduce the full ARMA structure into existing autoregressive attention mechanisms. By using an indirect MA weight generation method, we incorporate the MA term while maintaining the time complexity and parameter size of the underlying efficient attention models. We further explore how indirect parameter generation can produce implicit MA weights that align with the modeling requirements for local temporal impacts. Experimental results show that incorporating the ARMA structure consistently improves the performance of various AR attentions on TSF tasks, achieving state-of-the-art results.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4df609838f79c84e7e68d38c32072bc0d41efb09" target='_blank'>
-              Survey and Taxonomy: The Role of Data-Centric AI in Transformer-Based Time Series Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/38269db5cf2621d020be004caca5919c83acffa8" target='_blank'>
+              Autoregressive Moving-average Attention Mechanism for Time Series Forecasting
               </a>
             </td>
           <td>
-            Jingjing Xu, Caesar Wu, Yuan-Fang Li, Grégoire Danoy, Pascal Bouvry
+            Jiecheng Lu, Xu Han, Yan Sun, Shihao Yang
           </td>
-          <td>2024-07-29</td>
+          <td>2024-10-04</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>23</td>
+          <td>2</td>
         </tr>
 
-        <tr id="As the most representative scenario of spatial-temporal forecasting tasks, the traffic forecasting task attracted numerous attention from machine learning community due to its intricate correlation both in space and time dimension. Existing methods often treat road networks over time as spatial-temporal graphs, addressing spatial and temporal representations independently. However, these approaches struggle to capture the dynamic topology of road networks, encounter issues with message passing mechanisms and over-smoothing, and face challenges in learning spatial and temporal relationships separately. To address these limitations, we propose treating nodes in road networks at different time steps as independent spatial-temporal tokens and feeding them into a vanilla Transformer to learn complex spatial-temporal patterns, design STformer achieving SOTA. Given its quadratic complexity, we introduce a variant NSTformer based on Nystr$\ddot{o}$m method to approximate self-attention with linear complexity but even slightly better than former in a few cases astonishingly. Extensive experimental results on traffic datasets demonstrate that the proposed method achieves state-of-the-art performance at an affordable computational cost. Our code will be made available.">
+        <tr id="Time position embeddings capture the positional information of time steps, often serving as auxiliary inputs to enhance the predictive capabilities of time series models. However, existing models exhibit limitations in capturing intricate time positional information and effectively utilizing these embeddings. To address these limitations, this paper proposes a novel model called D2Vformer. Unlike typical prediction methods that rely on RNNs or Transformers, this approach can directly handle scenarios where the predicted sequence is not adjacent to the input sequence or where its length dynamically changes. In comparison to conventional methods, D2Vformer undoubtedly saves a significant amount of training resources. In D2Vformer, the Date2Vec module uses the timestamp information and feature sequences to generate time position embeddings. Afterward, D2Vformer introduces a new fusion block that utilizes an attention mechanism to explore the similarity in time positions between the embeddings of the input sequence and the predicted sequence, thereby generating predictions based on this similarity. Through extensive experiments on six datasets, we demonstrate that Date2Vec outperforms other time position embedding methods, and D2Vformer surpasses state-of-the-art methods in both fixed-length and variable-length prediction tasks.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/68b260739d1efb80558305eabc68f65b7a872e8b" target='_blank'>
-              Linear Attention is Enough in Spatial-Temporal Forecasting
+          <td><a href="https://www.semanticscholar.org/paper/319ef61a7e0685e57794aa31f0ecff8dfdfbafdf" target='_blank'>
+              D2Vformer: A Flexible Time Series Prediction Model Based on Time Position Embedding
               </a>
             </td>
           <td>
-            Xinyu Ning
+            Xiaobao Song, Hao Wang, Liwei Deng, Yuxin He, Wenming Cao, Andrew Chi-Sing Leung
           </td>
-          <td>2024-08-17</td>
+          <td>2024-09-17</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>1</td>
         </tr>
 
-        <tr id="Representing temporal-structured samples is essential for effective time series analysis tasks. So far, recurrent networks, convolution networks and transformer-style models have been successively applied in temporal data representation, yielding notable results. However, most existing methods primarily focus on modeling and representing the variation patterns within time series in the time domain. As a highly abstracted information entity, 1D time series couples various patterns such as trends, seasonality, and dramatic changes (instantaneous high dynamic), it is difficult to exploit these highly coupled properties merely by analysis tools on purely time domain. To this end, we present Spectrogram Analysis and Representation Network (SpecAR-Net). SpecAR-Net aims at learning more comprehensive representations by modeling raw time series in both time and frequency domain, where an efficient joint extraction of time-frequency features is achieved through a group of learnable 2D multi-scale parallel complex convolution blocks. Experimental results show that the SpecAR-Net achieves excellent performance on 5 major downstream tasks i.e., classification, anomaly detection, imputation, long- and short-term forecasting. Code and appendix are available at https://github.com/Dongyi2go/SpecAR_Net.">
+        <tr id="In many domains, such as healthcare, time-series data is often irregularly sampled with varying intervals between observations. This poses challenges for classical time-series models that require equally spaced data. To address this, we propose a novel time-series Transformer called Trajectory Generative Pre-trained Transformer (TrajGPT). TrajGPT employs a novel Selective Recurrent Attention (SRA) mechanism, which utilizes a data-dependent decay to adaptively filter out irrelevant past information based on contexts. By interpreting TrajGPT as discretized ordinary differential equations (ODEs), it effectively captures the underlying continuous dynamics and enables time-specific inference for forecasting arbitrary target timesteps. Experimental results demonstrate that TrajGPT excels in trajectory forecasting, drug usage prediction, and phenotype classification without requiring task-specific fine-tuning. By evolving the learned continuous dynamics, TrajGPT can interpolate and extrapolate disease risk trajectories from partially-observed time series. The visualization of predicted health trajectories shows that TrajGPT forecasts unseen diseases based on the history of clinically relevant phenotypes (i.e., contexts).">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/1dfaff572dd20e45c52824bd90981982d242c6dc" target='_blank'>
-              SpecAR-Net: Spectrogram Analysis and Representation Network for Time Series
+          <td><a href="https://www.semanticscholar.org/paper/1de2eba7172509c524a8adacc9179f5bae225bb1" target='_blank'>
+              TrajGPT: Irregular Time-Series Representation Learning for Health Trajectory Analysis
               </a>
             </td>
           <td>
-            Yi Dong, Liwen Zhang, Youcheng Zhang, Shi Peng, Wen Chen, Zhe Ma
+            Ziyang Song, Qingcheng Lu, He Zhu, David L. Buckeridge, Yue Li
           </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
-          <td>0</td>
+          <td>2024-10-03</td>
+          <td>ArXiv</td>
           <td>0</td>
+          <td>2</td>
         </tr>
 
-        <tr id="Learning useful representations for continuous-time dynamic graphs (CTDGs) is challenging, due to the concurrent need to span long node interaction histories and grasp nuanced temporal details. In particular, two problems emerge: (1) Encoding longer histories requires more computational resources, making it crucial for CTDG models to maintain low computational complexity to ensure efficiency; (2) Meanwhile, more powerful models are needed to identify and select the most critical temporal information within the extended context provided by longer histories. To address these problems, we propose a CTDG representation learning model named DyGMamba, originating from the popular Mamba state space model (SSM). DyGMamba first leverages a node-level SSM to encode the sequence of historical node interactions. Another time-level SSM is then employed to exploit the temporal patterns hidden in the historical graph, where its output is used to dynamically select the critical information from the interaction history. We validate DyGMamba experimentally on the dynamic link prediction task. The results show that our model achieves state-of-the-art in most cases. DyGMamba also maintains high efficiency in terms of computational resources, making it possible to capture long temporal dependencies with a limited computation budget.">
+        <tr id="Time series forecasting is extensively applied across diverse domains. Transformer-based models demonstrate significant potential in modeling cross-time and cross-variable interaction. However, we notice that the cross-variable correlation of multivariate time series demonstrates multifaceted (positive and negative correlations) and dynamic progression over time, which is not well captured by existing Transformer-based models. To address this issue, we propose a TimeCNN model to refine cross-variable interactions to enhance time series forecasting. Its key innovation is timepoint-independent, where each time point has an independent convolution kernel, allowing each time point to have its independent model to capture relationships among variables. This approach effectively handles both positive and negative correlations and adapts to the evolving nature of variable relationships over time. Extensive experiments conducted on 12 real-world datasets demonstrate that TimeCNN consistently outperforms state-of-the-art models. Notably, our model achieves significant reductions in computational requirements (approximately 60.46%) and parameter count (about 57.50%), while delivering inference speeds 3 to 4 times faster than the benchmark iTransformer model">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/aeb6fbef9a2f455a71773488e09f1c4f6686ac55" target='_blank'>
-              DyGMamba: Efficiently Modeling Long-Term Temporal Dependency on Continuous-Time Dynamic Graphs with State Space Models
+          <td><a href="https://www.semanticscholar.org/paper/70716b8b8c42001bd9093d3b115ac80b3f06da85" target='_blank'>
+              TimeCNN: Refining Cross-Variable Interaction on Time Point for Time Series Forecasting
               </a>
             </td>
           <td>
-            Zifeng Ding, Yifeng Li, Yuan He, Antonio Norelli, Jingcheng Wu, Volker Tresp, Yunpu Ma, Michael Bronstein
+            Ao Hu, Dongkai Wang, Yong Dai, Shiyi Qi, Liangjian Wen, Jun Wang, Zhi Chen, Xun Zhou, Zenglin Xu, Jiang Duan
           </td>
-          <td>2024-08-08</td>
+          <td>2024-10-07</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>22</td>
+          <td>3</td>
         </tr>
 
-        <tr id="The COVID-19 pandemic and influenza outbreaks have underscored the critical need for predictive models that can effectively integrate spatial and temporal dynamics to enable accurate epidemic forecasting. Traditional time-series analysis approaches have fallen short in capturing the intricate interplay between these factors. Recent advancements have witnessed the incorporation of graph neural networks and machine learning techniques to bridge this gap, enhancing predictive accuracy and providing novel insights into disease spread mechanisms. Notable endeavors include leveraging human mobility data, employing transfer learning, and integrating advanced models such as Transformers and Graph Convolutional Networks (GCNs) to improve forecasting performance across diverse geographies for both influenza and COVID-19. However, these models often face challenges related to data quality, model transferability, and potential overfitting, highlighting the necessity for more adaptable and robust approaches. This paper introduces the Graph Attention-based Spatial Temporal (GAST) model, which employs graph attention networks (GATs) to overcome these limitations by providing a nuanced understanding of epidemic dynamics through a sophisticated spatio-temporal analysis framework. Our contributions include the development and validation of the GAST model, demonstrating its superior forecasting capabilities for influenza and COVID-19 spread, with a particular focus on short-term, daily predictions. The model’s application to both influenza and COVID-19 datasets showcases its versatility and potential to inform public health interventions across a range of infectious diseases.">
+        <tr id="Long-term Time Series Forecasting (LTSF) is critical for numerous real-world applications, such as electricity consumption planning, financial forecasting, and disease propagation analysis. LTSF requires capturing long-range dependencies between inputs and outputs, which poses significant challenges due to complex temporal dynamics and high computational demands. While linear models reduce model complexity by employing frequency domain decomposition, current approaches often assume stationarity and filter out high-frequency components that may contain crucial short-term fluctuations. In this paper, we introduce MMFNet, a novel model designed to enhance long-term multivariate forecasting by leveraging a multi-scale masked frequency decomposition approach. MMFNet captures fine, intermediate, and coarse-grained temporal patterns by converting time series into frequency segments at varying scales while employing a learnable mask to filter out irrelevant components adaptively. Extensive experimentation with benchmark datasets shows that MMFNet not only addresses the limitations of the existing methods but also consistently achieves good performance. Specifically, MMFNet achieves up to 6.0% reductions in the Mean Squared Error (MSE) compared to state-of-the-art models designed for multivariate forecasting tasks.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/628bd28abb35127fc3d4e794dfc8e0860b738427" target='_blank'>
-              Modeling epidemic dynamics using Graph Attention based Spatial Temporal networks
+          <td><a href="https://www.semanticscholar.org/paper/3f9c427b47cc457b438f61dd74255b6839302655" target='_blank'>
+              MMFNet: Multi-Scale Frequency Masking Neural Network for Multivariate Time Series Forecasting
               </a>
             </td>
           <td>
-            Xiaofeng Zhu, Yi Zhang, Haoru Ying, Huanning Chi, Guanqun Sun, Lingxia Zeng
+            Aitian Ma, Dongsheng Luo, Mo Sha
           </td>
-          <td>2024-07-15</td>
-          <td>PLOS ONE</td>
-          <td>0</td>
+          <td>2024-10-02</td>
+          <td>ArXiv</td>
           <td>0</td>
+          <td>2</td>
         </tr>
 
-        <tr id="The stock market is a fundamental component of financial systems, reflecting economic health, providing investment opportunities, and influencing global dynamics. Accurate stock market predictions can lead to significant gains and promote better investment decisions. However, predicting stock market trends is challenging due to their non-linear and stochastic nature. This study investigates the efficacy of advanced deep learning models for short-term trend forecasting using daily and hourly closing prices from the S&P 500 index and the Brazilian ETF EWZ. The models explored include Temporal Convolutional Networks (TCN), Neural Basis Expansion Analysis for Time Series Forecasting (N-BEATS), Temporal Fusion Transformers (TFT), Neural Hierarchical Interpolation for Time Series Forecasting (N-HiTS), and Time-series Dense Encoder (TiDE). Furthermore, we introduce the Extended Long Short-Term Memory for Time Series (xLSTM-TS) model, an xLSTM adaptation optimised for time series prediction. Wavelet denoising techniques were applied to smooth the signal and reduce minor fluctuations, providing cleaner data as input for all approaches. Denoising significantly improved performance in predicting stock price direction. Among the models tested, xLSTM-TS consistently outperformed others. For example, it achieved a test accuracy of 72.82% and an F1 score of 73.16% on the EWZ daily dataset. By leveraging advanced deep learning models and effective data preprocessing techniques, this research provides valuable insights into the application of machine learning for market movement forecasting, highlighting both the potential and the challenges involved.">
+        <tr id="Time series~(TS) modeling is essential in dynamic systems like weather prediction and anomaly detection. Recent studies utilize Large Language Models (LLMs) for TS modeling, leveraging their powerful pattern recognition capabilities. These methods primarily position LLMs as the predictive backbone, often omitting the mathematical modeling within traditional TS models, such as periodicity. However, disregarding the potential of LLMs also overlooks their pattern recognition capabilities. To address this gap, we introduce \textit{LLM-TS Integrator}, a novel framework that effectively integrates the capabilities of LLMs into traditional TS modeling. Central to this integration is our \textit{mutual information} module. The core of this \textit{mutual information} module is a traditional TS model enhanced with LLM-derived insights for improved predictive abilities. This enhancement is achieved by maximizing the mutual information between traditional model's TS representations and LLM's textual representation counterparts, bridging the two modalities. Moreover, we recognize that samples vary in importance for two losses: traditional prediction and mutual information maximization. To address this variability, we introduce the \textit{sample reweighting} module to improve information utilization. This module assigns dual weights to each sample: one for prediction loss and another for mutual information loss, dynamically optimizing these weights via bi-level optimization. Our method achieves state-of-the-art or comparable performance across five mainstream TS tasks, including short-term and long-term forecasting, imputation, classification, and anomaly detection.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/bbcd75d3d798abef451ae501b4954b9878a57d40" target='_blank'>
-              An Evaluation of Deep Learning Models for Stock Market Trend Prediction
+          <td><a href="https://www.semanticscholar.org/paper/013fefa7c76d4f07cd5b40bb12553cb17be9e98c" target='_blank'>
+              LLM-TS Integrator: Integrating LLM for Enhanced Time Series Modeling
               </a>
             </td>
           <td>
-            Gonzalo Lopez Gil, Paul Duhamel-Sebline, Andrew McCarren
+            Can Chen, Gabriel Oliveira, H. S. Noghabi, Tristan Sylvain
           </td>
-          <td>2024-08-22</td>
+          <td>2024-10-21</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>0</td>
+          <td>2</td>
         </tr>
 
-        <tr id="Time series forecasting has essential applications across various domains. For instance, forecasting power time series can optimize energy usage and bolster grid stability and reliability. Existing models based on transformer architecture are limited to classical design, ignoring the impact of spatial information and noise on model architecture design. Therefore, we propose an atypical design of Transformer-based models for multivariate time series forecasting. This design consists of two critical components: (i) spectral clustering center of time series employed as the focal point for attention computation; (ii) alternating attention mechanism wherein each query transformer is compatible with spectral clustering centers, executing attention at the sequence level instead of the token level. The alternating design has a two-fold benefit: firstly, it eliminates the uncertainty noise present in the dependent variable sequence of the channel input, and secondly, it incorporates the Euclidean distance to mitigate the impact of extreme values on the attention matrix, thereby aligning predictions more closely to the sequence's natural progression. Experiments on ten real-world datasets, encompassing Wind, Electricity, Weather, and others, demonstrate that our Spectral Central Alternating Transformer (SCAT) outperforms state-of-the-art methods (SOTA) by an average of 42.8% in prediction performance in power time series forecasting.">
+        <tr id="Recent research has challenged the necessity of complex deep learning architectures for time series forecasting, demonstrating that simple linear models can often outperform sophisticated approaches. Building upon this insight, we introduce a novel architecture the Temporal Linear Net (TLN), that extends the capabilities of linear models while maintaining interpretability and computational efficiency. TLN is designed to effectively capture both temporal and feature-wise dependencies in multivariate time series data. Our approach is a variant of TSMixer that maintains strict linearity throughout its architecture. TSMixer removes activation functions, introduces specialized kernel initializations, and incorporates dilated convolutions to handle various time scales, while preserving the linear nature of the model. Unlike transformer-based models that may lose temporal information due to their permutation-invariant nature, TLN explicitly preserves and leverages the temporal structure of the input data. A key innovation of TLN is its ability to compute an equivalent linear model, offering a level of interpretability not found in more complex architectures such as TSMixer. This feature allows for seamless conversion between the full TLN model and its linear equivalent, facilitating both training flexibility and inference optimization.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/f0478468dc09d88ac952d50786c548b2bee25e6a" target='_blank'>
-              SCAT: A Time Series Forecasting with Spectral Central Alternating Transformers
+          <td><a href="https://www.semanticscholar.org/paper/e21eecfcf3e1b0d611d383f980313b51108d1e5a" target='_blank'>
+              A Temporal Linear Network for Time Series Forecasting
               </a>
             </td>
           <td>
-            Chengjie Zhou, Chao Che, Pengfei Wang, Qiang Zhang
+            Remi Genet, Hugo Inzirillo
           </td>
-          <td>2024-08-01</td>
-          <td>Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence</td>
+          <td>2024-10-28</td>
+          <td>ArXiv</td>
           <td>0</td>
-          <td>8</td>
+          <td>2</td>
         </tr>
 
     </tbody>
diff --git a/site/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378/index.html b/site/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378/index.html
index 063c6836..7ed0f4c5 100644
--- a/site/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378/index.html
+++ b/site/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:58 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:45 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,7 +484,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-03-15</td>
           <td>PLOS Computational Biology</td>
-          <td>3</td>
+          <td>5</td>
           <td>55</td>
         </tr>
 
@@ -544,8 +544,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-06-01</td>
           <td>Nonlinear Dynamics</td>
-          <td>29</td>
-          <td>91</td>
+          <td>30</td>
+          <td>92</td>
         </tr>
 
         <tr id="Systems biology is a new discipline built upon the premise that an understanding of how cells and organisms carry out their functions cannot be gained by looking at cellular components in isolation. Instead, consideration of the interplay between the parts of systems is indispensable for analyzing, modeling, and predicting systems' behavior. Studying biological processes under this premise, systems biology combines experimental techniques and computational methods in order to construct predictive models. Both in building and utilizing models of biological systems, inverse problems arise at several occasions, for example, (i) when experimental time series and steady state data are used to construct biochemical reaction networks, (ii) when model parameters are identified that capture underlying mechanisms or (iii) when desired qualitative behavior such as bistability or limit cycle oscillations is engineered by proper choices of parameter combinations. In this paper we review principles of the modeling process in systems biology and illustrate the ill-posedness and regularization of parameter identification problems in that context. Furthermore, we discuss the methodology of qualitative inverse problems and demonstrate how sparsity enforcing regularization allows the determination of key reaction mechanisms underlying the qualitative behavior.">
@@ -560,6 +560,18 @@ <h3 data-intro='Recommendations for the article'>
           <td>48</td>
         </tr>
 
+        <tr id="Biological systems exhibit complex dynamics that differential equations can often adeptly represent. Ordinary differential equation models are widespread; until recently their construction has required extensive prior knowledge of the system. Machine learning methods offer alternative means of model construction: differential equation models can be learnt from data via model discovery using sparse identification of nonlinear dynamics (SINDy). However, SINDy struggles with realistic levels of biological noise and does not incorporate prior knowledge of the system. We propose a data-driven framework for model discovery and model selection using hybrid dynamical systems: partial models containing missing terms. Neural networks are used to approximate the unknown dynamics of a system, enabling the denoising the data while simultaneously learning the latent dynamics. Simulations from the fitted neural network are then used to infer models using SINDy. We show, via model selection, that model discovery in SINDy with hybrid dynamical systems outperforms alternative approaches. We find it possible to infer models correctly up to high levels of biological noise of different types. We demonstrate the potential to learn models from sparse, noisy data in application to a canonical cell state transition using data derived from single-cell transcriptomics. Overall, this approach provides a practical framework for model discovery in biology in cases where data are noisy and sparse, of particular utility when the underlying biological mechanisms are partially but incompletely known.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/596f63cbf87cf14a0e7859c17058a82b15c760ef" target='_blank'>Data-driven model discovery and model selection for noisy biological systems</a></td>
+          <td>
+            Xiaojun Wu, MeiLu McDermott, Adam L. Maclean
+          </td>
+          <td>2024-10-04</td>
+          <td>bioRxiv</td>
+          <td>0</td>
+          <td>19</td>
+        </tr>
+
         <tr id="None">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/1626b462b65d4084a24fb31c4e4ca3fc212a307a" target='_blank'>Set-base dynamical parameter estimation and model invalidation for biochemical reaction networks</a></td>
@@ -569,7 +581,19 @@ <h3 data-intro='Recommendations for the article'>
           <td>2010-05-25</td>
           <td>BMC Systems Biology</td>
           <td>67</td>
-          <td>47</td>
+          <td>48</td>
+        </tr>
+
+        <tr id="The combination of machine learning (ML) and sparsity-promoting techniques is enabling direct extraction of governing equations from data, revolutionizing computational modeling in diverse fields of science and engineering. The discovered dynamical models could be used to address challenges in climate science, neuroscience, ecology, finance, epidemiology, and beyond. However, most existing sparse identification methods for discovering dynamical systems treat the whole system as one without considering the interactions between subsystems. As a result, such models are not able to capture small changes in the emergent system behavior. To address this issue, we developed a new method called Sparse Identification of Nonlinear Dynamical Systems from Graph-structured data (SINDyG), which incorporates the network structure into sparse regression to identify model parameters that explain the underlying network dynamics. SINDyG discovers the governing equations of network dynamics while offering improvements in accuracy and model simplicity.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/4da518ae886702aed27abdd8f94fea3042ab9377" target='_blank'>Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems</a></td>
+          <td>
+            Mohammad Amin Basiri, Sina Khanmohammadi
+          </td>
+          <td>2024-09-02</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>3</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/0acd117521ef5aafb09fed02ab415523b330b058/index.html b/site/recommendations/0acd117521ef5aafb09fed02ab415523b330b058/index.html
index c8590302..26d380e6 100644
--- a/site/recommendations/0acd117521ef5aafb09fed02ab415523b330b058/index.html
+++ b/site/recommendations/0acd117521ef5aafb09fed02ab415523b330b058/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:06 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:51 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,7 +484,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-11-01</td>
           <td>Chaos</td>
-          <td>3</td>
+          <td>4</td>
           <td>11</td>
         </tr>
 
@@ -508,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
 
         <tr id="None">
@@ -520,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
 
@@ -532,8 +532,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>250</td>
-          <td>12</td>
+          <td>289</td>
+          <td>13</td>
         </tr>
 
         <tr id="Discovering the partial differential equations underlying spatio-temporal datasets from very limited and highly noisy observations is of paramount interest in many scientific fields. However, it remains an open question to know when model discovery algorithms based on sparse regression can actually recover the underlying physical processes. In this work, we show the design matrices used to infer the equations by sparse regression can violate the irrepresentability condition (IRC) of the Lasso, even when derived from analytical PDE solutions (i.e. without additional noise). Sparse regression techniques which can recover the true underlying model under violated IRC conditions are therefore required, leading to the introduction of the randomised adaptive Lasso. We show once the latter is integrated within the deep learning model discovery framework DeepMod, a wide variety of nonlinear and chaotic canonical PDEs can be recovered: (1) up to $\mathcal{O}(2)$ higher noise-to-sample ratios than state-of-the-art algorithms, (2) with a single set of hyperparameters, which paves the road towards truly automated model discovery.">
@@ -545,7 +545,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2021-06-22</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
 
         <tr id="We investigate the problem of learning an evolution equation directly from some given data. This work develops a learning algorithm to identify the terms in the underlying partial differential equations and to approximate the coefficients of the terms only using data. The algorithm uses sparse optimization in order to perform feature selection and parameter estimation. The features are data driven in the sense that they are constructed using nonlinear algebraic equations on the spatial derivatives of the data. Several numerical experiments show the proposed method's robustness to data noise and size, its ability to capture the true features of the data, and its capability of performing additional analytics. Examples include shock equations, pattern formation, fluid flow and turbulence, and oscillatory convection.">
@@ -557,7 +557,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2017-01-16</td>
           <td></td>
           <td>0</td>
-          <td>15</td>
+          <td>16</td>
         </tr>
 
         <tr id="We propose a regression method based upon group sparsity that is capable of discovering parametrized governing dynamical equations of motion of a given system by time series measurements. The method balances model complexity and regression accuracy by selecting a parsimonious model via Pareto analysis. This gives a promising new technique for disambiguating governing equations from simple parametric dependencies in physical, biological and engineering systems.">
@@ -569,7 +569,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2017-12-01</td>
           <td>2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)</td>
           <td>12</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b/index.html b/site/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b/index.html
index 75402715..85f2f0b4 100644
--- a/site/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b/index.html
+++ b/site/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:15 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:10 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,7 +484,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-10-24</td>
           <td>ArXiv</td>
-          <td>6</td>
+          <td>9</td>
           <td>50</td>
         </tr>
 
@@ -508,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>54</td>
+          <td>1</td>
+          <td>55</td>
         </tr>
 
         <tr id="Accurate forecasting of multivariate time series is an extensively studied subject in finance, transportation, and computer science. Fully mining the correlation and causation between the variables in a multivariate time series exhibits noticeable results in improving the performance of a time series model. Recently, some models have explored the dependencies between variables through end-to-end graph structure learning without the need for predefined graphs. However, current models do not incorporate the trade-off between efficiency and flexibility and lack the guidance of domain knowledge in the design of graph structure learning algorithms. This paper alleviates the above issues by proposing Balanced Graph Structure Learning for Forecasting (BGSLF), a novel deep learning model that joins graph structure learning and forecasting. Technically, BGSLF leverages the spatial information into convolutional operations and extracts temporal dynamics using the diffusion convolutional recurrent network. The proposed framework balance the trade-off between efficiency and flexibility by introducing Multi-Graph Generation Network (MGN) and Graph Selection Module. In addition, a method named Smooth Sparse Unit (SSU) is designed to sparse the learned graph structures, which conforms to the sparse spatial correlations in the real world. Extensive experiments on four real-world datasets demonstrate that our model achieves state-of-the-art performances with minor trainable parameters. Code will be made publicly available.">
@@ -521,7 +521,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2022-01-24</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>10</td>
+          <td>11</td>
         </tr>
 
         <tr id="We study a recent class of models which uses graph neural networks (GNNs) to improve forecasting in multivariate time series. The core assumption behind these models is that there is a latent graph between the time series (nodes) that governs the evolution of the multivariate time series. By parameterizing a graph in a differentiable way, the models aim to improve forecasting quality. We compare four recent models of this class on the forecasting task. Further, we perform ablations to study their behavior under changing conditions, e.g., when disabling the graph-learning modules and providing the ground-truth relations instead. Based on our findings, we propose novel ways of combining the existing architectures.">
@@ -533,7 +533,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2021-09-10</td>
           <td>ArXiv</td>
           <td>12</td>
-          <td>46</td>
+          <td>47</td>
         </tr>
 
         <tr id="Multivariate time-series forecasting is a critical task for many applications, and graph time-series network is widely studied due to its capability to capture the spatial-temporal correlation simultaneously. However, most existing works focus more on learning with the explicit prior graph structure, while ignoring potential information from the implicit graph structure, yielding incomplete structure modeling. Some recent works attempts to learn the intrinsic or implicit graph structure directly, while lacking a way to combine explicit prior structure with implicit structure together. In this paper, we propose Regularized Graph Structure Learning (RGSL) model to incorporate both explicit prior structure and implicit structure together, and learn the forecasting deep networks along with the graph structure. RGSL consists of two innovative modules. First, we derive an implicit dense similarity matrix through node embedding, and learn the sparse graph structure using the Regularized Graph Generation (RGG) based on the Gumbel Softmax trick. Second, we propose a Laplacian Matrix Mixed-up Module (LM3) to fuse the explicit graph and implicit graph together. We conduct experiments on three real-word datasets. Results show that the proposed RGSL model outperforms existing graph forecasting algorithms with a notable margin, while learning meaningful graph structure simultaneously. Our code and models are made publicly available at https://github.com/alipay/RGSL.git.">
@@ -544,8 +544,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-07-01</td>
           <td>ArXiv, DBLP</td>
-          <td>40</td>
-          <td>34</td>
+          <td>44</td>
+          <td>35</td>
         </tr>
 
         <tr id="Time series forecasting is an extensively studied subject in statistics, economics, and computer science. Exploration of the correlation and causation among the variables in a multivariate time series shows promise in enhancing the performance of a time series model. When using deep neural networks as forecasting models, we hypothesize that exploiting the pairwise information among multiple (multivariate) time series also improves their forecast. If an explicit graph structure is known, graph neural networks (GNNs) have been demonstrated as powerful tools to exploit the structure. In this work, we propose learning the structure simultaneously with the GNN if the graph is unknown. We cast the problem as learning a probabilistic graph model through optimizing the mean performance over the graph distribution. The distribution is parameterized by a neural network so that discrete graphs can be sampled differentiably through reparameterization. Empirical evaluations show that our method is simpler, more efficient, and better performing than a recently proposed bilevel learning approach for graph structure learning, as well as a broad array of forecasting models, either deep or non-deep learning based, and graph or non-graph based.">
@@ -556,8 +556,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-01-18</td>
           <td>ArXiv</td>
-          <td>179</td>
-          <td>37</td>
+          <td>199</td>
+          <td>38</td>
         </tr>
 
         <tr id="Multi-variate time series forecasting is an important problem with a wide range of applications. Recent works model the relations between time-series as graphs and have shown that propagating information over the relation graph can improve time series forecasting. However, in many cases, relational information is not available or is noisy and reliable. Moreover, most works ignore the underlying uncertainty of time-series both for structure learning and deriving the forecasts resulting in the structure not capturing the uncertainty resulting in forecast distributions with poor uncertainty estimates. We tackle this challenge and introduce STOIC, that leverages stochastic correlations between time-series to learn underlying structure between time-series and to provide well-calibrated and accurate forecasts. Over a wide-range of benchmark datasets STOIC provides around 16% more accurate and 14% better-calibrated forecasts. STOIC also shows better adaptation to noise in data during inference and captures important and useful relational information in various benchmarks.">
diff --git a/site/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b/index.html b/site/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b/index.html
index 7a8f8335..c7c39fca 100644
--- a/site/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b/index.html
+++ b/site/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:17 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:12 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -478,41 +478,41 @@ <h3 data-intro='Recommendations for the article'>
 
         <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4fb78450650894091b0a55b5504a8b0a6f3dec37" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
           <td>
             Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>28</td>
+          <td>69</td>
           <td>18</td>
         </tr>
 
-        <tr id="This research examines the use of Large Language Models (LLMs) in predicting time series, with a specific focus on the LLMTIME model. Despite the established effectiveness of LLMs in tasks such as text generation, language translation, and sentiment analysis, this study highlights the key challenges that large language models encounter in the context of time series prediction. We assess the performance of LLMTIME across multiple datasets and introduce classical almost periodic functions as time series to gauge its effectiveness. The empirical results indicate that while large language models can perform well in zero-shot forecasting for certain datasets, their predictive accuracy diminishes notably when confronted with diverse time series data and traditional signals. The primary finding of this study is that the predictive capacity of LLMTIME, similar to other LLMs, significantly deteriorates when dealing with time series data that contain both periodic and trend components, as well as when the signal comprises complex frequency components.">
+        <tr id="Foundation models of time series have not been fully developed due to the limited availability of time series corpora and the underexploration of scalable pre-training. Based on the similar sequential formulation of time series and natural language, increasing research demonstrates the feasibility of leveraging large language models (LLM) for time series. Nevertheless, the inherent autoregressive property and decoder-only architecture of LLMs have not been fully considered, resulting in insufficient utilization of LLM abilities. To fully revitalize the general-purpose token transition and multi-step generation capability of large language models, we propose AutoTimes to repurpose LLMs as autoregressive time series forecasters, which projects time series into the embedding space of language tokens and autoregressively generates future predictions with arbitrary lengths. Compatible with any decoder-only LLMs, the consequent forecaster exhibits the flexibility of the lookback length and scalability with larger LLMs. Further, we formulate time series as prompts, extending the context for prediction beyond the lookback window, termed in-context forecasting. By introducing LLM-embedded textual timestamps, AutoTimes can utilize chronological information to align multivariate time series. Empirically, AutoTimes achieves state-of-the-art with 0.1% trainable parameters and over $5\times$ training/inference speedup compared to advanced LLM-based forecasters. Code is available at this repository: https://github.com/thuml/AutoTimes.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/11e794835a1f7d46560666d90c8eab3c24d84c05" target='_blank'>An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/d488445bb2bf6719bc48a4d39bd906116274abda" target='_blank'>AutoTimes: Autoregressive Time Series Forecasters via Large Language Models</a></td>
           <td>
-            Rui Cao, Qiao Wang
+            Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long
           </td>
-          <td>2024-08-09</td>
+          <td>2024-02-04</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>3</td>
+          <td>5</td>
+          <td>67</td>
         </tr>
 
-        <tr id="Foundation models of time series have not been fully developed due to the limited availability of time series corpora and the underexploration of scalable pre-training. Based on the similar sequential formulation of time series and natural language, increasing research demonstrates the feasibility of leveraging large language models (LLM) for time series. Nevertheless, the inherent autoregressive property and decoder-only architecture of LLMs have not been fully considered, resulting in insufficient utilization of LLM abilities. To further exploit the general-purpose token transition and multi-step generation ability of large language models, we propose AutoTimes to repurpose LLMs as autoregressive time series forecasters, which independently projects time series segments into the embedding space and autoregressively generates future predictions with arbitrary lengths. Compatible with any decoder-only LLMs, the consequent forecaster exhibits the flexibility of the lookback length and scalability of the LLM size. Further, we formulate time series as prompts, extending the context for prediction beyond the lookback window, termed in-context forecasting. By adopting textual timestamps as position embeddings, AutoTimes integrates multimodality for multivariate scenarios. Empirically, AutoTimes achieves state-of-the-art with 0.1% trainable parameters and over 5 times training/inference speedup compared to advanced LLM-based forecasters.">
+        <tr id="This research examines the use of Large Language Models (LLMs) in predicting time series, with a specific focus on the LLMTIME model. Despite the established effectiveness of LLMs in tasks such as text generation, language translation, and sentiment analysis, this study highlights the key challenges that large language models encounter in the context of time series prediction. We assess the performance of LLMTIME across multiple datasets and introduce classical almost periodic functions as time series to gauge its effectiveness. The empirical results indicate that while large language models can perform well in zero-shot forecasting for certain datasets, their predictive accuracy diminishes notably when confronted with diverse time series data and traditional signals. The primary finding of this study is that the predictive capacity of LLMTIME, similar to other LLMs, significantly deteriorates when dealing with time series data that contain both periodic and trend components, as well as when the signal comprises complex frequency components.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/d488445bb2bf6719bc48a4d39bd906116274abda" target='_blank'>AutoTimes: Autoregressive Time Series Forecasters via Large Language Models</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/11e794835a1f7d46560666d90c8eab3c24d84c05" target='_blank'>An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting</a></td>
           <td>
-            Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long
+            Rui Cao, Qiao Wang
           </td>
-          <td>2024-02-04</td>
+          <td>2024-08-09</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>66</td>
+          <td>0</td>
+          <td>3</td>
         </tr>
 
-        <tr id="Large language models (LLMs) are being applied to time series tasks, particularly time series forecasting. However, are language models actually useful for time series? After a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade the forecasting results -- in most cases the results even improved. We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and reveal that patching and attention structures perform similarly to state-of-the-art LLM-based forecasters.">
+        <tr id="Large language models (LLMs) are being applied to time series forecasting. But are language models actually useful for time series? In a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade forecasting performance -- in most cases, the results even improve! We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and find that patching and attention structures perform similarly to LLM-based forecasters.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/df0d604b8e8e3b2947d9865d735f204c08635012" target='_blank'>Are Language Models Actually Useful for Time Series Forecasting?</a></td>
           <td>
@@ -520,8 +520,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-06-22</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>1</td>
+          <td>13</td>
+          <td>3</td>
         </tr>
 
         <tr id="Deep learning has contributed remarkably to the advancement of time series analysis. Still, deep models can encounter performance bottlenecks in real-world data-scarce scenarios, which can be concealed due to the performance saturation with small models on current benchmarks. Meanwhile, large models have demonstrated great powers in these scenarios through large-scale pre-training. Continuous progress has been achieved with the emergence of large language models, exhibiting unprecedented abilities such as few-shot generalization, scalability, and task generality, which are however absent in small deep models. To change the status quo of training scenario-specific small models from scratch, this paper aims at the early development of large time series models (LTSM). During pre-training, we curate large-scale datasets with up to 1 billion time points, unify heterogeneous time series into single-series sequence (S3) format, and develop the GPT-style architecture toward LTSMs. To meet diverse application needs, we convert forecasting, imputation, and anomaly detection of time series into a unified generative task. The outcome of this study is a Time Series Transformer (Timer), which is generative pre-trained by next token prediction and adapted to various downstream tasks with promising capabilities as an LTSM. Code and datasets are available at: https://github.com/thuml/Large-Time-Series-Model.">
@@ -532,8 +532,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>4</td>
-          <td>66</td>
+          <td>21</td>
+          <td>67</td>
         </tr>
 
         <tr id="In this paper, we introduce TimeGPT, the first foundation model for time series, capable of generating accurate predictions for diverse datasets not seen during training. We evaluate our pre-trained model against established statistical, machine learning, and deep learning methods, demonstrating that TimeGPT zero-shot inference excels in performance, efficiency, and simplicity. Our study provides compelling evidence that insights from other domains of artificial intelligence can be effectively applied to time series analysis. We conclude that large-scale time series models offer an exciting opportunity to democratize access to precise predictions and reduce uncertainty by leveraging the capabilities of contemporary advancements in deep learning.">
@@ -544,8 +544,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-10-05</td>
           <td>ArXiv</td>
-          <td>43</td>
-          <td>1</td>
+          <td>60</td>
+          <td>5</td>
         </tr>
 
         <tr id="Predicting future values in multivariate time series is vital across various domains. This work explores the use of large language models (LLMs) for this task. However, LLMs typically handle one-dimensional data. We introduce MultiCast, a zero-shot LLM-based approach for multivariate time series forecasting. It allows LLMs to receive multivariate time series as input, through three novel token multiplexing solutions that effectively reduce dimensionality while preserving key repetitive patterns. Additionally, a quantization scheme helps LLMs to better learn these patterns, while significantly reducing token use for practical applications. We showcase the performance of our approach in terms of RMSE and execution time against state-of-the-art approaches on three real-world datasets.">
@@ -568,8 +568,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-16</td>
           <td>ArXiv</td>
-          <td>3</td>
-          <td>6</td>
+          <td>5</td>
+          <td>8</td>
         </tr>
 
         <tr id="Time series forecasting holds significant importance in many real-world dynamic systems and has been extensively studied. Unlike natural language process (NLP) and computer vision (CV), where a single large model can tackle multiple tasks, models for time series forecasting are often specialized, necessitating distinct designs for different tasks and applications. While pre-trained foundation models have made impressive strides in NLP and CV, their development in time series domains has been constrained by data sparsity. Recent studies have revealed that large language models (LLMs) possess robust pattern recognition and reasoning abilities over complex sequences of tokens. However, the challenge remains in effectively aligning the modalities of time series data and natural language to leverage these capabilities. In this work, we present Time-LLM, a reprogramming framework to repurpose LLMs for general time series forecasting with the backbone language models kept intact. We begin by reprogramming the input time series with text prototypes before feeding it into the frozen LLM to align the two modalities. To augment the LLM's ability to reason with time series data, we propose Prompt-as-Prefix (PaP), which enriches the input context and directs the transformation of reprogrammed input patches. The transformed time series patches from the LLM are finally projected to obtain the forecasts. Our comprehensive evaluations demonstrate that Time-LLM is a powerful time series learner that outperforms state-of-the-art, specialized forecasting models. Moreover, Time-LLM excels in both few-shot and zero-shot learning scenarios.">
@@ -580,7 +580,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-10-03</td>
           <td>ArXiv</td>
-          <td>134</td>
+          <td>195</td>
           <td>9</td>
         </tr>
 
diff --git a/site/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277/index.html b/site/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277/index.html
index f30b1a5c..c7fb2726 100644
--- a/site/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277/index.html
+++ b/site/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:30 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:18 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -485,7 +485,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2024-05-23</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>6</td>
+          <td>7</td>
         </tr>
 
         <tr id="Large language models (LLMs) have been applied in many fields and have developed rapidly in recent years. As a classic machine learning task, time series forecasting has recently been boosted by LLMs. Recent works treat large language models as \emph{zero-shot} time series reasoners without further fine-tuning, which achieves remarkable performance. However, there are some unexplored research problems when applying LLMs for time series forecasting under the zero-shot setting. For instance, the LLMs' preferences for the input time series are less understood. In this paper, by comparing LLMs with traditional time series forecasting models, we observe many interesting properties of LLMs in the context of time series forecasting. First, our study shows that LLMs perform well in predicting time series with clear patterns and trends, but face challenges with datasets lacking periodicity. This observation can be explained by the ability of LLMs to recognize the underlying period within datasets, which is supported by our experiments. In addition, the input strategy is investigated, and it is found that incorporating external knowledge and adopting natural language paraphrases substantially improve the predictive performance of LLMs for time series. Overall, our study contributes insight into LLMs' advantages and limitations in time series forecasting under different conditions.">
@@ -496,8 +496,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-16</td>
           <td>ArXiv</td>
-          <td>3</td>
-          <td>6</td>
+          <td>5</td>
+          <td>8</td>
         </tr>
 
         <tr id="Time-series forecasting (TSF) finds broad applications in real-world scenarios. Prompting off-the-shelf Large Language Models (LLMs) demonstrates strong zero-shot TSF capabilities while preserving computational efficiency. However, existing prompting methods oversimplify TSF as language next-token predictions, overlooking its dynamic nature and lack of integration with state-of-the-art prompt strategies such as Chain-of-Thought. Thus, we propose LSTPrompt, a novel approach for prompting LLMs in zero-shot TSF tasks. LSTPrompt decomposes TSF into short-term and long-term forecasting sub-tasks, tailoring prompts to each. LSTPrompt guides LLMs to regularly reassess forecasting mechanisms to enhance adaptability. Extensive evaluations demonstrate consistently better performance of LSTPrompt than existing prompting methods, and competitive results compared to foundation TSF models.">
@@ -508,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-25</td>
           <td>ArXiv</td>
-          <td>6</td>
-          <td>7</td>
+          <td>11</td>
+          <td>8</td>
         </tr>
 
         <tr id="This paper presents a new perspective on time series forecasting. In existing time series forecasting methods, the models take a sequence of numerical values as input and yield numerical values as output. The existing SOTA models are largely based on the Transformer architecture, modified with multiple encoding mechanisms to incorporate the context and semantics around the historical data. Inspired by the successes of pre-trained language foundation models, we pose a question about whether these models can also be adapted to solve time-series forecasting. Thus, we propose a new forecasting paradigm: prompt-based time series forecasting (PromptCast). In this novel task, the numerical input and output are transformed into prompts and the forecasting task is framed in a sentence-to-sentence manner, making it possible to directly apply language models for forecasting purposes. To support and facilitate the research of this task, we also present a large-scale dataset (PISA) that includes three real-world forecasting scenarios. We evaluate different SOTA numerical-based forecasting methods and language generation models. The benchmark results with various forecasting settings demonstrate the proposed PromptCast with language generation models is a promising research direction. Additionally, in comparison to conventional numerical-based forecasting, PromptCast shows a much better generalization ability under the zero-shot setting.">
@@ -520,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-09-20</td>
           <td>IEEE Transactions on Knowledge and Data Engineering</td>
-          <td>63</td>
+          <td>80</td>
           <td>17</td>
         </tr>
 
@@ -533,7 +533,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2024-06-20</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>23</td>
+          <td>24</td>
         </tr>
 
         <tr id="Time series data are ubiquitous across various domains, making time series analysis critically important. Traditional time series models are task-specific, featuring singular functionality and limited generalization capacity. Recently, large language foundation models have unveiled their remarkable capabilities for cross-task transferability, zero-shot/few-shot learning, and decision-making explainability. This success has sparked interest in the exploration of foundation models to solve multiple time series challenges simultaneously. There are two main research lines, namely pre-training foundation models from scratch for time series and adapting large language foundation models for time series. They both contribute to the development of a unified model that is highly generalizable, versatile, and comprehensible for time series analysis. This survey offers a 3E analytical framework for comprehensive examination of related research. Specifically, we examine existing works from three dimensions, namely Effectiveness, Efficiency and Explainability. In each dimension, we focus on discussing how related works devise tailored solution by considering unique challenges in the realm of time series. Furthermore, we provide a domain taxonomy to help followers keep up with the domain-specific advancements. In addition, we introduce extensive resources to facilitate the field's development, including datasets, open-source, time series libraries. A GitHub repository is also maintained for resource updates (https://github.com/start2020/Awesome-TimeSeries-LLM-FM).">
@@ -544,7 +544,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-05-03</td>
           <td>ArXiv</td>
-          <td>4</td>
+          <td>7</td>
           <td>47</td>
         </tr>
 
@@ -556,8 +556,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-05-23</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>1</td>
+          <td>2</td>
+          <td>2</td>
         </tr>
 
         <tr id="Multivariate time-series forecasting is vital in various domains, e.g., economic planning and weather prediction. Deep train-from-scratch models have exhibited effective performance yet require large amounts of data, which limits real-world applicability. Recently, researchers have leveraged the representation learning transferability of pre-trained Large Language Models (LLMs) to handle limited non-linguistic datasets effectively. However, incorporating LLMs with time-series data presents challenges of limited adaptation due to different compositions between time-series and linguistic data, and the inability to process multi-scale temporal information. To tackle these challenges, we propose LLM4TS, a framework for time-series forecasting with pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the \textit{time-series alignment} stage to align LLMs with the nuances of time-series data, and the \textit{forecasting fine-tuning} stage for downstream time-series forecasting tasks. Furthermore, our framework features a novel two-level aggregation method that integrates multi-scale temporal data within pre-trained LLMs, enhancing their ability to interpret time-specific information. In experiments across 7 time-series forecasting datasets, LLM4TS is superior to existing state-of-the-art methods compared with trained-from-scratch models in full-shot scenarios, and also achieves an average improvement of 6.84% in MSE in few-shot scenarios. In addition, evaluations compared with different self-supervised learning approaches highlight LLM4TS's effectiveness with representation learning in forecasting tasks.">
@@ -568,32 +568,20 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-08-16</td>
           <td>ArXiv</td>
-          <td>16</td>
-          <td>2</td>
+          <td>22</td>
+          <td>3</td>
         </tr>
 
-        <tr id="Large language models (LLMs) are being applied to time series tasks, particularly time series forecasting. However, are language models actually useful for time series? After a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade the forecasting results -- in most cases the results even improved. We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and reveal that patching and attention structures perform similarly to state-of-the-art LLM-based forecasters.">
+        <tr id="Multi-modal large language models (MLLMs) have enabled numerous advances in understanding and reasoning in domains like vision, but we have not yet seen this broad success for time-series. Although prior works on time-series MLLMs have shown promising performance in time-series forecasting, very few works show how an LLM could be used for time-series reasoning in natural language. We propose a novel multi-modal time-series LLM approach that learns generalizable information across various domains with powerful zero-shot performance. First, we train a lightweight time-series encoder on top of an LLM to directly extract time-series information. Then, we fine-tune our model with chain-of-thought augmented time-series tasks to encourage the model to generate reasoning paths. We show that our model learns a latent representation that reflects specific time-series features (e.g. slope, frequency), as well as outperforming GPT-4o on a set of zero-shot reasoning tasks on a variety of domains.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/df0d604b8e8e3b2947d9865d735f204c08635012" target='_blank'>Are Language Models Actually Useful for Time Series Forecasting?</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/e2b1ba7e73ca4ce0abef8c87f2c266e502b3b7b4" target='_blank'>Towards Time Series Reasoning with LLMs</a></td>
           <td>
-            Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen
+            Winnie Chow, Lauren Gardiner, Haraldur T. Hallgr'imsson, Maxwell A. Xu, Shirley You Ren
           </td>
-          <td>2024-06-22</td>
+          <td>2024-09-17</td>
           <td>ArXiv</td>
+          <td>2</td>
           <td>1</td>
-          <td>1</td>
-        </tr>
-
-        <tr id="Motivated by recent advances in large language models for Natural Language Processing (NLP), we design a time-series foundation model for forecasting whose out-of-the-box zero-shot performance on a variety of public datasets comes close to the accuracy of state-of-the-art supervised forecasting models for each individual dataset. Our model is based on pretraining a patched-decoder style attention model on a large time-series corpus, and can work well across different forecasting history lengths, prediction lengths and temporal granularities.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/f45f85fa1beaa795c24c4ff86f1f2deece72252f" target='_blank'>A decoder-only foundation model for time-series forecasting</a></td>
-          <td>
-            Abhimanyu Das, Weihao Kong, Rajat Sen, Yichen Zhou
-          </td>
-          <td>2023-10-14</td>
-          <td>ArXiv</td>
-          <td>51</td>
-          <td>14</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f/index.html b/site/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f/index.html
index b43fea3f..380005ad 100644
--- a/site/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f/index.html
+++ b/site/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:17 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:13 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-05-22</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>2</td>
+          <td>1</td>
+          <td>6</td>
         </tr>
 
         <tr id="Graph Neural Networks (GNNs) have shown promising potential in graph representation learning. The majority of GNNs define a local message-passing mechanism, propagating information over the graph by stacking multiple layers. These methods, however, are known to suffer from two major limitations: over-squashing and poor capturing of long-range dependencies. Recently, Graph Transformers (GTs) emerged as a powerful alternative to Message-Passing Neural Networks (MPNNs). GTs, however, have quadratic computational cost, lack inductive biases on graph structures, and rely on complex Positional/Structural Encodings (SE/PE). In this paper, we show that while Transformers, complex message-passing, and SE/PE are sufficient for good performance in practice, neither is necessary. Motivated by the recent success of State Space Models (SSMs), such as Mamba, we present Graph Mamba Networks (GMNs), a general framework for a new class of GNNs based on selective SSMs. We discuss and categorize the new challenges when adapting SSMs to graph-structured data, and present four required and one optional steps to design GMNs, where we choose (1) Neighborhood Tokenization, (2) Token Ordering, (3) Architecture of Bidirectional Selective SSM Encoder, (4) Local Encoding, and dispensable (5) PE and SE. We further provide theoretical justification for the power of GMNs. Experiments demonstrate that despite much less computational cost, GMNs attain an outstanding performance in long-range, small-scale, large-scale, and heterophilic benchmark datasets.">
@@ -496,11 +496,11 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-13</td>
           <td>ArXiv, DBLP</td>
-          <td>33</td>
+          <td>39</td>
           <td>9</td>
         </tr>
 
-        <tr id="Message-passing graph neural networks (GNNs), while excelling at capturing local relationships, often struggle with long-range dependencies on graphs. Conversely, graph transformers (GTs) enable information exchange between all nodes but oversimplify the graph structure by treating them as a set of fixed-length vectors. This work proposes a novel architecture, NeuralWalker, that overcomes the limitations of both methods by combining random walks with message passing. NeuralWalker achieves this by treating random walks as sequences, allowing for the application of recent advances in sequence models in order to capture long-range dependencies within these walks. Based on this concept, we propose a framework that offers (1) more expressive graph representations through random walk sequences, (2) the ability to utilize any sequence model for capturing long-range dependencies, and (3) the flexibility by integrating various GNN and GT architectures. Our experimental evaluations demonstrate that NeuralWalker achieves significant performance improvements on 19 graph and node benchmark datasets, notably outperforming existing methods by up to 13% on the PascalVoc-SP and COCO-SP datasets. Code is available at https://github.com/BorgwardtLab/NeuralWalker.">
+        <tr id="Message-passing graph neural networks (GNNs) excel at capturing local relationships but struggle with long-range dependencies in graphs. In contrast, graph transformers (GTs) enable global information exchange but often oversimplify the graph structure by representing graphs as sets of fixed-length vectors. This work introduces a novel architecture that overcomes the shortcomings of both approaches by combining the long-range information of random walks with local message passing. By treating random walks as sequences, our architecture leverages recent advances in sequence models to effectively capture long-range dependencies within these walks. Based on this concept, we propose a framework that offers (1) more expressive graph representations through random walk sequences, (2) the ability to utilize any sequence model for capturing long-range dependencies, and (3) the flexibility by integrating various GNN and GT architectures. Our experimental evaluations demonstrate that our approach achieves significant performance improvements on 19 graph and node benchmark datasets, notably outperforming existing methods by up to 13\% on the PascalVoc-SP and COCO-SP datasets. The code is available at https://github.com/BorgwardtLab/NeuralWalker.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/0702ac9db3ea20810ffdb9f35bde85cdfaf27a3d" target='_blank'>Learning Long Range Dependencies on Graphs via Random Walks</a></td>
           <td>
@@ -508,11 +508,11 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-06-05</td>
           <td>ArXiv</td>
-          <td>0</td>
+          <td>2</td>
           <td>8</td>
         </tr>
 
-        <tr id="Machine learning on graphs has recently found extensive applications across domains. However, the commonly used Message Passing Neural Networks (MPNNs) suffer from limited expressive power and struggle to capture long-range dependencies. Graph transformers offer a strong alternative due to their global attention mechanism, but they come with great computational overheads, especially for large graphs. In recent years, State Space Models (SSMs) have emerged as a compelling approach to replace full attention in transformers to model sequential data. It blends the strengths of RNNs and CNNs, offering a) efficient computation, b) the ability to capture long-range dependencies, and c) good generalization across sequences of various lengths. However, extending SSMs to graph-structured data presents unique challenges due to the lack of canonical node ordering in graphs. In this work, we propose Graph State Space Convolution (GSSC) as a principled extension of SSMs to graph-structured data. By leveraging global permutation-equivariant set aggregation and factorizable graph kernels that rely on relative node distances as the convolution kernels, GSSC preserves all three advantages of SSMs. We demonstrate the provably stronger expressiveness of GSSC than MPNNs in counting graph substructures and show its effectiveness across 10 real-world, widely used benchmark datasets, where GSSC achieves best results on 7 out of 10 datasets with all significant improvements compared to the state-of-the-art baselines and second-best results on the other 3 datasets. Our findings highlight the potential of GSSC as a powerful and scalable model for graph machine learning. Our code is available at https://github.com/Graph-COM/GSSC.">
+        <tr id="Machine learning on graphs has recently found extensive applications across domains. However, the commonly used Message Passing Neural Networks (MPNNs) suffer from limited expressive power and struggle to capture long-range dependencies. Graph transformers offer a strong alternative due to their global attention mechanism, but they come with great computational overheads, especially for large graphs. In recent years, State Space Models (SSMs) have emerged as a compelling approach to replace full attention in transformers to model sequential data. It blends the strengths of RNNs and CNNs, offering a) efficient computation, b) the ability to capture long-range dependencies, and c) good generalization across sequences of various lengths. However, extending SSMs to graph-structured data presents unique challenges due to the lack of canonical node ordering in graphs. In this work, we propose Graph State Space Convolution (GSSC) as a principled extension of SSMs to graph-structured data. By leveraging global permutation-equivariant set aggregation and factorizable graph kernels that rely on relative node distances as the convolution kernels, GSSC preserves all three advantages of SSMs. We demonstrate the provably stronger expressiveness of GSSC than MPNNs in counting graph substructures and show its effectiveness across 11 real-world, widely used benchmark datasets. GSSC achieves the best results on 6 out of 11 datasets with all significant improvements compared to the state-of-the-art baselines and second-best results on the other 5 datasets. Our findings highlight the potential of GSSC as a powerful and scalable model for graph machine learning. Our code is available at https://github.com/Graph-COM/GSSC.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/93a9b7940e902f989e27f94061d3b699a2fc4d13" target='_blank'>What Can We Learn from State Space Models for Machine Learning on Graphs?</a></td>
           <td>
@@ -524,6 +524,18 @@ <h3 data-intro='Recommendations for the article'>
           <td>3</td>
         </tr>
 
+        <tr id="Graph representation learning (GRL) is fundamental in multi-graph applications like molecular property prediction. Graph neural networks (GNNs) have emerged as a popular method for GRL. However, existing GRL methods primarily focus on designing GNN models with enhanced expressiveness while overlooking the memory efficiency of algorithms during training. The memory inefficiency problem is caused by a contextual constraint imposed on node representations, which requires each node to be context-dependent on its input graph. In this paper, we propose a novel method, called context sketching (COS), for memory-efficient graph representation learning in multi-graph scenarios. We first formally define the contextual constraint based on the enclosed $\infty-$hop subgraphs of nodes. Subsequently, we propose to relax the original contextual constraint by requiring each node to be context-dependent on its enclosed $k-$hop subgraph $(k \ll \infty)$ which is a contextual sketch of the enclosed $\infty$-hop subgraph. Lastly, we prove that COS constructs an optimal solution to a memory-related objective associated with graph coarsening. Experiments on four widely used benchmark datasets demonstrate that COS can reduce the memory footprint of baselines by a large margin with almost no accuracy loss.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/54288b7e9ce29a338f2b11981c6b84f4f0cb9a50" target='_blank'>Context Sketching for Memory-efficient Graph Representation Learning</a></td>
+          <td>
+            Kai-Lang Yao, Wusuo Li
+          </td>
+          <td>2023-12-01</td>
+          <td>2023 IEEE International Conference on Data Mining (ICDM)</td>
+          <td>0</td>
+          <td>4</td>
+        </tr>
+
         <tr id="Graph Neural Networks (GNNs) have shown impressive performance in graph representation learning, but they face challenges in capturing long-range dependencies due to their limited expressive power. To address this, Graph Transformers (GTs) were introduced, utilizing self-attention mechanism to effectively model pairwise node relationships. Despite their advantages, GTs suffer from quadratic complexity w.r.t. the number of nodes in the graph, hindering their applicability to large graphs. In this work, we present Graph-Enhanced Contextual Operator (GECO), a scalable and effective alternative to GTs that leverages neighborhood propagation and global convolutions to effectively capture local and global dependencies in quasilinear time. Our study on synthetic datasets reveals that GECO reaches 169x speedup on a graph with 2M nodes w.r.t. optimized attention. Further evaluations on diverse range of benchmarks showcase that GECO scales to large graphs where traditional GTs often face memory and time limitations. Notably, GECO consistently achieves comparable or superior quality compared to baselines, improving the SOTA up to 4.5%, and offering a scalable and effective solution for large-scale graph learning.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/48052c9ebe066b9fcd653897dabf18582ec7e7fb" target='_blank'>A Scalable and Effective Alternative to Graph Transformers</a></td>
@@ -544,7 +556,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-10-08</td>
           <td>ArXiv</td>
-          <td>59</td>
+          <td>63</td>
           <td>19</td>
         </tr>
 
@@ -560,18 +572,6 @@ <h3 data-intro='Recommendations for the article'>
           <td>28</td>
         </tr>
 
-        <tr id="Recent works have demonstrated the benefits of capturing long-distance dependency in graphs by deeper graph neural networks (GNNs). But deeper GNNs suffer from the long-lasting scalability challenge due to the neighborhood explosion problem in large-scale graphs. In this work, we propose to capture long-distance dependency in graphs by shallower models instead of deeper models, which leads to a much more efficient model, LazyGNN, for graph representation learning. Moreover, we demonstrate that LazyGNN is compatible with existing scalable approaches (such as sampling methods) for further accelerations through the development of mini-batch LazyGNN. Comprehensive experiments demonstrate its superior prediction performance and scalability on large-scale benchmarks. The implementation of LazyGNN is available at https://github.com/RXPHD/Lazy_GNN.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/5ca04cde78542b8d58c4c5b2db91ddd65b56d917" target='_blank'>LazyGNN: Large-Scale Graph Neural Networks via Lazy Propagation</a></td>
-          <td>
-            Rui Xue, Haoyu Han, MohamadAli Torkamani, Jian Pei, Xiaorui Liu
-          </td>
-          <td>2023-02-03</td>
-          <td>ArXiv</td>
-          <td>9</td>
-          <td>7</td>
-        </tr>
-
   </tbody>
   <tfoot>
     <tr>
diff --git a/site/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68/index.html b/site/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68/index.html
index b786038b..538a404e 100644
--- a/site/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68/index.html
+++ b/site/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:34 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:28 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -483,8 +483,8 @@ <h3 data-intro='Recommendations for the article'>
             Andreas Doerr, Christian Daniel, Martin Schiegg, D. Nguyen-Tuong, S. Schaal, Marc Toussaint, Sebastian Trimpe
           </td>
           <td>2018-01-31</td>
-          <td>ArXiv, DBLP, MAG</td>
-          <td>111</td>
+          <td>ArXiv, MAG, DBLP</td>
+          <td>113</td>
           <td>93</td>
         </tr>
 
@@ -500,6 +500,18 @@ <h3 data-intro='Recommendations for the article'>
           <td>40</td>
         </tr>
 
+        <tr id="World modelling is essential for understanding and predicting the dynamics of complex systems by learning both spatial and temporal dependencies. However, current frameworks, such as Transformers and selective state-space models like Mambas, exhibit limitations in efficiently encoding spatial and temporal structures, particularly in scenarios requiring long-term high-dimensional sequence modelling. To address these issues, we propose a novel recurrent framework, the \textbf{FACT}ored \textbf{S}tate-space (\textbf{FACTS}) model, for spatial-temporal world modelling. The FACTS framework constructs a graph-structured memory with a routing mechanism that learns permutable memory representations, ensuring invariance to input permutations while adapting through selective state-space propagation. Furthermore, FACTS supports parallel computation of high-dimensional sequences. We empirically evaluate FACTS across diverse tasks, including multivariate time series forecasting and object-centric world modelling, demonstrating that it consistently outperforms or matches specialised state-of-the-art models, despite its general-purpose world modelling design.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/ec840755867d3f5cf175cb57de963f042297f4ef" target='_blank'>FACTS: A Factored State-Space Framework For World Modelling</a></td>
+          <td>
+            Li Nanbo, Firas Laakom, Yucheng Xu, Wenyi Wang, Jurgen Schmidhuber
+          </td>
+          <td>2024-10-28</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>7</td>
+        </tr>
+
         <tr id="Time series with long-term structure arise in a variety of contexts and capturing this temporal structure is a critical challenge in time series analysis for both inference and forecasting settings. Traditionally, state space models have been successful in providing uncertainty estimates of trajectories in the latent space. More recently, deep learning, attention-based approaches have achieved state of the art performance for sequence modeling, though often require large amounts of data and parameters to do so. We propose Stanza, a nonlinear, non-stationary state space model as an intermediate approach to fill the gap between traditional models and modern deep learning approaches for complex time series. Stanza strikes a balance between competitive forecasting accuracy and probabilistic, interpretable inference for highly structured time series. In particular, Stanza achieves forecasting accuracy competitive with deep LSTMs on real-world datasets, especially for multi-step ahead forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/9c619e58d073772ff298228d47afcab625c7f37a" target='_blank'>Stanza: A Nonlinear State Space Model for Probabilistic Inference in Non-Stationary Time Series</a></td>
@@ -509,7 +521,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2020-06-11</td>
           <td>ArXiv</td>
           <td>6</td>
-          <td>31</td>
+          <td>32</td>
         </tr>
 
         <tr id="Outstanding achievements of graph neural networks for spatiotemporal time series analysis show that relational constraints introduce an effective inductive bias into neural forecasting architectures. Often, however, the relational information characterizing the underlying data-generating process is unavailable and the practitioner is left with the problem of inferring from data which relational graph to use in the subsequent processing stages. We propose novel, principled - yet practical - probabilistic score-based methods that learn the relational dependencies as distributions over graphs while maximizing end-to-end the performance at task. The proposed graph learning framework is based on consolidated variance reduction techniques for Monte Carlo score-based gradient estimation, is theoretically grounded, and, as we show, effective in practice. In this paper, we focus on the time series forecasting problem and show that, by tailoring the gradient estimators to the graph learning problem, we are able to achieve state-of-the-art performance while controlling the sparsity of the learned graph and the computational scalability. We empirically assess the effectiveness of the proposed method on synthetic and real-world benchmarks, showing that the proposed solution can be used as a stand-alone graph identification procedure as well as a graph learning component of an end-to-end forecasting architecture.">
@@ -520,7 +532,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-05-26</td>
           <td>J. Mach. Learn. Res.</td>
-          <td>11</td>
+          <td>14</td>
           <td>50</td>
         </tr>
 
@@ -528,7 +540,7 @@ <h3 data-intro='Recommendations for the article'>
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/05cd3afc8208f1c0dd61b09a90f35dd42497e175" target='_blank'>Multi-Knowledge Fusion Network for Time Series Representation Learning</a></td>
           <td>
-            Sagar Srinivas Sakhinana, Shivam Gupta, Krishna Sai Sudhir Aripirala, Venkataramana Runkana
+            Sakhinana Sagar Srinivas, Shivam Gupta, Krishna Sai Sudhir Aripirala, Venkataramana Runkana
           </td>
           <td>2024-08-22</td>
           <td>ArXiv</td>
@@ -536,44 +548,28 @@ <h3 data-intro='Recommendations for the article'>
           <td>3</td>
         </tr>
 
-        <tr id="Over the past few years, research on deep graph learning has shifted from static graphs to temporal graphs in response to real-world complex systems that exhibit dynamic behaviors. In practice, temporal graphs are formalized as an ordered sequence of static graph snapshots observed at discrete time points. Sequence models such as RNNs or Transformers have long been the predominant backbone networks for modeling such temporal graphs. Yet, despite the promising results, RNNs struggle with long-range dependencies, while transformers are burdened by quadratic computational complexity. Recently, state space models (SSMs), which are framed as discretized representations of an underlying continuous-time linear dynamical system, have garnered substantial attention and achieved breakthrough advancements in independent sequence modeling. In this work, we undertake a principled investigation that extends SSM theory to temporal graphs by integrating structural information into the online approximation objective via the adoption of a Laplacian regularization term. The emergent continuous-time system introduces novel algorithmic challenges, thereby necessitating our development of GraphSSM, a graph state space model for modeling the dynamics of temporal graphs. Extensive experimental results demonstrate the effectiveness of our GraphSSM framework across various temporal graph benchmarks.">
+        <tr id="Real-world dynamical systems often consist of multiple stochastic subsystems that interact with each other. Modeling and forecasting the behavior of such dynamics are generally not easy, due to the inherent hardness in understanding the complicated interactions and evolutions of their constituents. This paper introduces the relational state-space model (R-SSM), a sequential hierarchical latent variable model that makes use of graph neural networks (GNNs) to simulate the joint state transitions of multiple correlated objects. By letting GNNs cooperate with SSM, R-SSM provides a flexible way to incorporate relational information into the modeling of multi-object dynamics. We further suggest augmenting the model with normalizing flows instantiated for vertex-indexed random variables and propose two auxiliary contrastive objectives to facilitate the learning. The utility of R-SSM is empirically evaluated on synthetic and real time series datasets.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/919e5db29c7b7be4468b975eb4c0fa4a543165fc" target='_blank'>State Space Models on Temporal Graphs: A First-Principles Study</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/7a1e5377b08489c2969f73c56efc557e34f578e1" target='_blank'>Relational State-Space Model for Stochastic Multi-Object Systems</a></td>
           <td>
-            Jintang Li, Ruofan Wu, Xinzhou Jin, Boqun Ma, Liang Chen, Zibin Zheng
+            Fan Yang, Ling Chen, Fan Zhou, Yusong Gao, Wei Cao
           </td>
-          <td>2024-06-03</td>
+          <td>2020-01-13</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>11</td>
+          <td>8</td>
+          <td>60</td>
         </tr>
 
-        <tr id="Time series modeling is a well-established problem, which often requires that methods (1) expressively represent complicated dependencies, (2) forecast long horizons, and (3) efficiently train over long sequences. State-space models (SSMs) are classical models for time series, and prior works combine SSMs with deep learning layers for efficient sequence modeling. However, we find fundamental limitations with these prior approaches, proving their SSM representations cannot express autoregressive time series processes. We thus introduce SpaceTime, a new state-space time series architecture that improves all three criteria. For expressivity, we propose a new SSM parameterization based on the companion matrix -- a canonical representation for discrete-time processes -- which enables SpaceTime's SSM layers to learn desirable autoregressive processes. For long horizon forecasting, we introduce a"closed-loop"variation of the companion SSM, which enables SpaceTime to predict many future time-steps by generating its own layer-wise inputs. For efficient training and inference, we introduce an algorithm that reduces the memory and compute of a forward pass with the companion matrix. With sequence length $\ell$ and state-space size $d$, we go from $\tilde{O}(d \ell)$ na\"ively to $\tilde{O}(d + \ell)$. In experiments, our contributions lead to state-of-the-art results on extensive and diverse benchmarks, with best or second-best AUROC on 6 / 7 ECG and speech time series classification, and best MSE on 14 / 16 Informer forecasting tasks. Furthermore, we find SpaceTime (1) fits AR($p$) processes that prior deep SSMs fail on, (2) forecasts notably more accurately on longer horizons than prior state-of-the-art, and (3) speeds up training on real-world ETTh1 data by 73% and 80% relative wall-clock time over Transformers and LSTMs.">
+        <tr id="Over the past few years, research on deep graph learning has shifted from static graphs to temporal graphs in response to real-world complex systems that exhibit dynamic behaviors. In practice, temporal graphs are formalized as an ordered sequence of static graph snapshots observed at discrete time points. Sequence models such as RNNs or Transformers have long been the predominant backbone networks for modeling such temporal graphs. Yet, despite the promising results, RNNs struggle with long-range dependencies, while transformers are burdened by quadratic computational complexity. Recently, state space models (SSMs), which are framed as discretized representations of an underlying continuous-time linear dynamical system, have garnered substantial attention and achieved breakthrough advancements in independent sequence modeling. In this work, we undertake a principled investigation that extends SSM theory to temporal graphs by integrating structural information into the online approximation objective via the adoption of a Laplacian regularization term. The emergent continuous-time system introduces novel algorithmic challenges, thereby necessitating our development of GraphSSM, a graph state space model for modeling the dynamics of temporal graphs. Extensive experimental results demonstrate the effectiveness of our GraphSSM framework across various temporal graph benchmarks.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/a7d68b1702af08ce4dbbf2cd0b083e744ae5c6be" target='_blank'>Effectively Modeling Time Series with Simple Discrete State Spaces</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/919e5db29c7b7be4468b975eb4c0fa4a543165fc" target='_blank'>State Space Models on Temporal Graphs: A First-Principles Study</a></td>
           <td>
-            Michael Zhang, Khaled Kamal Saab, Michael Poli, Tri Dao, Karan Goel, Christopher Ré
+            Jintang Li, Ruofan Wu, Xinzhou Jin, Boqun Ma, Liang Chen, Zibin Zheng
           </td>
-          <td>2023-03-16</td>
+          <td>2024-06-03</td>
           <td>ArXiv</td>
-          <td>30</td>
-          <td>44</td>
-        </tr>
-
-        <tr id="
-
- Gaussian state space models have been used for decades as generative models of sequential data. They admit an intuitive probabilistic interpretation, have a simple functional form, and enjoy widespread adoption. We introduce a unified algorithm to efficiently learn a broad class of linear and non-linear state space models, including variants where the emission and transition distributions are modeled by deep neural networks. Our learning algorithm simultaneously learns a compiled inference network and the generative model, leveraging a structured variational approximation parameterized by recurrent neural networks to mimic the posterior distribution. We apply the learning algorithm to both synthetic and real-world datasets, demonstrating its scalability and versatility. We find that using the structured approximation to the posterior results in models with significantly higher held-out likelihood.
-
-">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2af17f153e3fd71e15db9216b972aef222f46617" target='_blank'>Structured Inference Networks for Nonlinear State Space Models</a></td>
-          <td>
-            R. G. Krishnan, Uri Shalit, D. Sontag
-          </td>
-          <td>2016-09-30</td>
-          <td>ArXiv, DBLP, MAG</td>
-          <td>429</td>
-          <td>48</td>
+          <td>1</td>
+          <td>12</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196/index.html b/site/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196/index.html
index 78c607ca..fcd4156d 100644
--- a/site/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196/index.html
+++ b/site/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:04 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:47 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-10-18</td>
           <td>ArXiv</td>
-          <td>18</td>
-          <td>50</td>
+          <td>21</td>
+          <td>51</td>
         </tr>
 
         <tr id="In this paper we propose a model reduction and identification approach for multilinear dynamical system (MLDS) driven by noise. Compared to standard linear dynamical system based approaches which fit vector or matrix models to tensor time series, MLDS provides more natural, compact and accurate representation of tensorial data with fewer model parameters. The proposed algorithm for identifying MLDS parameters employs techniques from multilinear subspace learning: mulilinear Principal Component Analysis and multilinear regression. In addition compact array normal distribution is used to represent and estimate model error and output noise.We illustrate the benefits of the proposed approach on some real world datasets.">
@@ -520,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-11-22</td>
           <td>Biophysical Reviews</td>
-          <td>35</td>
+          <td>36</td>
           <td>5</td>
         </tr>
 
@@ -556,7 +556,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-07-02</td>
           <td>ArXiv</td>
-          <td>129</td>
+          <td>136</td>
           <td>23</td>
         </tr>
 
@@ -572,6 +572,18 @@ <h3 data-intro='Recommendations for the article'>
           <td>4</td>
         </tr>
 
+        <tr id="None">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/a9a3238808d20c5236a7c5f912c20e6dd729f88c" target='_blank'>Data-driven Reconstruction of Nonlinear Dynamics from Sparse Observation</a></td>
+          <td>
+            K. Yeo
+          </td>
+          <td>2019-06-10</td>
+          <td>J. Comput. Phys.</td>
+          <td>23</td>
+          <td>19</td>
+        </tr>
+
         <tr id="In recent years, we have witnessed a significant shift toward ever-more complex and ever-larger-scale systems in the majority of the grand societal challenges tackled in applied sciences. The need to comprehend and predict the dynamics of complex systems have spurred developments in large-scale simulations and a multitude of methods across several disciplines. The goals of understanding and prediction in complex dynamical systems, however, have been hindered by high dimensionality, complexity and chaotic behaviours. Recent advances in data-driven techniques and machine-learning approaches have revolutionized how we model and analyse complex systems. The integration of these techniques with dynamical systems theory opens up opportunities to tackle previously unattainable challenges in modelling and prediction of dynamical systems. While data-driven prediction methods have made great strides in recent years, it is still necessary to develop new techniques to improve their applicability to a wider range of complex systems in science and engineering. This focus issue shares recent developments in the field of complex dynamical systems with emphasis on data-driven, data-assisted and artificial intelligence-based discovery of dynamical systems. This article is part of the theme issue 'Data-driven prediction in dynamical systems'.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/e5aeec169d2c4d5e8e730a3c382006ba983b8709" target='_blank'>Data-driven prediction in dynamical systems: recent developments</a></td>
@@ -580,22 +592,10 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-06-20</td>
           <td>Philosophical transactions. Series A, Mathematical, physical, and engineering sciences</td>
-          <td>43</td>
+          <td>48</td>
           <td>33</td>
         </tr>
 
-        <tr id="This article discusses a newly developed online manifold learning method, subspace iteration using reduced models (SIRM), for the dimensionality reduction of dynamical systems. This method may be viewed as subspace iteration combined with a model reduction procedure. Specifically, starting with a test solution, the method solves a reduced model to obtain a more precise solution, and it repeats this process until sufficient accuracy is achieved. The reduced model is obtained by projecting the full model onto a subspace that is spanned by the dominant modes of an extended data ensemble. The extended data ensemble in this article contains not only the state vectors of some snapshots of the approximate solution from the previous iteration but also the associated tangent vectors. Therefore, the proposed manifold learning method takes advantage of the information of the original dynamical system to reduce the dynamics. Moreover, the learning procedure is computed in the online stage, as opposed to being compute...">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/418ebf87dea83c922735e9083fe533efc8a04b7f" target='_blank'>An Online Manifold Learning Approach for Model Reduction of Dynamical Systems</a></td>
-          <td>
-            Liqian Peng, K. Mohseni
-          </td>
-          <td>2012-10-10</td>
-          <td>SIAM J. Numer. Anal.</td>
-          <td>22</td>
-          <td>40</td>
-        </tr>
-
   </tbody>
   <tfoot>
     <tr>
diff --git a/site/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220/index.html b/site/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220/index.html
index c7318c3c..18f35bea 100644
--- a/site/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220/index.html
+++ b/site/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:05 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:50 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2019-06-25</td>
           <td>IEEE Access</td>
-          <td>98</td>
-          <td>65</td>
+          <td>103</td>
+          <td>67</td>
         </tr>
 
         <tr id="Energy conservation is a basic physics principle, the breakdown of which often implies new physics. This paper presents a method for data-driven "new physics" discovery. Specifically, given a trajectory governed by unknown forces, our neural new-physics detector (NNPhD) aims to detect new physics by decomposing the force field into conservative and nonconservative components, which are represented by a Lagrangian neural network (LNN) and an unconstrained neural network, respectively, trained to minimize the force recovery error plus a constant λ times the magnitude of the predicted nonconservative force. We show that a phase transition occurs at λ=1, universally for arbitrary forces. We demonstrate that NNPhD successfully discovers new physics in toy numerical experiments, rediscovering friction (1493) from a damped double pendulum, Neptune from Uranus' orbit (1846), and gravitational waves (2017) from an inspiraling orbit. We also show how NNPhD coupled with an integrator outperforms both an LNN and an unconstrained neural network for predicting the future of a damped double pendulum.">
@@ -496,8 +496,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-05-31</td>
           <td>Physical review. E</td>
-          <td>14</td>
-          <td>82</td>
+          <td>15</td>
+          <td>83</td>
         </tr>
 
         <tr id="Invariants and conservation laws convey critical information about the underlying dynamics of a system, yet it is generally infeasible to find them from large-scale data without any prior knowledge or human insight. We propose ConservNet to achieve this goal, a neural network that spontaneously discovers a conserved quantity from grouped data where the members of each group share invariants, similar to a general experimental setting where trajectories from different trials are observed. As a neural network trained with a novel and intuitive loss function called noise-variance loss, ConservNet learns the hidden invariants in each group of multi-dimensional observables in a data-driven, end-to-end manner. Our model successfully discovers underlying invariants from the simulated systems having invariants as well as a real-world double pendulum trajectory. Since the model is robust to various noises and data conditions compared to baseline, our approach is directly applicable to experimental data for discovering hidden conservation laws and further, general relationships between variables.">
@@ -532,7 +532,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-05-31</td>
           <td>2023 American Control Conference (ACC)</td>
-          <td>20</td>
+          <td>22</td>
           <td>38</td>
         </tr>
 
@@ -556,8 +556,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-02-17</td>
           <td>Communications Physics</td>
-          <td>23</td>
-          <td>30</td>
+          <td>25</td>
+          <td>31</td>
         </tr>
 
         <tr id="We present AI Poincaré, a machine learning algorithm for autodiscovering conserved quantities using trajectory data from unknown dynamical systems. We test it on five Hamiltonian systems, including the gravitational three-body problem, and find that it discovers not only all exactly conserved quantities, but also periodic orbits, phase transitions, and breakdown timescales for approximate conservation laws.">
@@ -568,7 +568,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-05-06</td>
           <td>Physical review letters</td>
-          <td>87</td>
+          <td>93</td>
           <td>82</td>
         </tr>
 
@@ -580,7 +580,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-09-01</td>
           <td>Nature Computational Science</td>
-          <td>0</td>
+          <td>1</td>
           <td>17</td>
         </tr>
 
@@ -592,8 +592,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-07-06</td>
           <td>ArXiv</td>
-          <td>23</td>
-          <td>27</td>
+          <td>25</td>
+          <td>28</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/3c9961153493370500020c81527b3548c96f81e0/index.html b/site/recommendations/3c9961153493370500020c81527b3548c96f81e0/index.html
index ec4d8c44..5dc3e7e9 100644
--- a/site/recommendations/3c9961153493370500020c81527b3548c96f81e0/index.html
+++ b/site/recommendations/3c9961153493370500020c81527b3548c96f81e0/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:55 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:42 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
 
         <tr id="None">
@@ -496,8 +496,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>250</td>
-          <td>12</td>
+          <td>289</td>
+          <td>13</td>
         </tr>
 
         <tr id="Harnessing data to discover the underlying governing laws or equations that describe the behavior of complex physical systems can significantly advance our modeling, simulation and understanding of such systems in various science and engineering disciplines. Recent advances in sparse identification show encouraging success in distilling closed-form governing equations from data for a wide range of nonlinear dynamical systems. However, the fundamental bottleneck of this approach lies in the robustness and scalability with respect to data scarcity and noise. This work introduces a novel physics-informed deep learning framework to discover governing partial differential equations (PDEs) from scarce and noisy data for nonlinear spatiotemporal systems. In particular, this approach seamlessly integrates the strengths of deep neural networks for rich representation learning, automatic differentiation and sparse regression to approximate the solution of system variables, compute essential derivatives, as well as identify the key derivative terms and parameters that form the structure and explicit expression of the PDEs. The efficacy and robustness of this method are demonstrated on discovering a variety of PDE systems with different levels of data scarcity and noise. The resulting computational framework shows the potential for closed-form model discovery in practical applications where large and accurate datasets are intractable to capture.">
@@ -509,7 +509,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2020-05-05</td>
           <td>ArXiv</td>
           <td>19</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
 
         <tr id="None">
@@ -520,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-11-22</td>
           <td>Biophysical Reviews</td>
-          <td>35</td>
+          <td>36</td>
           <td>5</td>
         </tr>
 
@@ -532,8 +532,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-11-19</td>
           <td>Proceedings of the Royal Society A</td>
-          <td>12</td>
-          <td>31</td>
+          <td>13</td>
+          <td>32</td>
         </tr>
 
         <tr id="Identifying governing equations for a dynamical system is a topic of critical interest across an array of disciplines, from mathematics to engineering to biology. Machine learning-specifically deep learning-techniques have shown their capabilities in approximating dynamics from data, but a shortcoming of traditional deep learning is that there is little insight into the underlying mapping beyond its numerical output for a given input. This limits their utility in analysis beyond simple prediction. Simultaneously, a number of strategies exist which identify models based on a fixed dictionary of basis functions, but most either require some intuition or insight about the system, or are susceptible to overfitting or a lack of parsimony. Here, we present a novel approach that combines the flexibility and accuracy of deep learning approaches with the utility of symbolic solutions: a deep neural network that generates a symbolic expression for the governing equations. We first describe the architecture for our model and then show the accuracy of our algorithm across a range of classical dynamical systems.">
@@ -544,7 +544,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-08-01</td>
           <td>Chaos</td>
-          <td>2</td>
+          <td>4</td>
           <td>37</td>
         </tr>
 
@@ -556,7 +556,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
 
@@ -568,8 +568,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-07-22</td>
           <td>Communications Physics</td>
-          <td>17</td>
-          <td>94</td>
+          <td>20</td>
+          <td>95</td>
         </tr>
 
         <tr id="Estimating governing equations from observed time-series data is crucial for understanding dynamical systems. From the perspective of system comprehension, the demand for accurate estimation and interpretable results has been particularly emphasized. Herein, we propose a novel data-driven method for estimating the governing equations of dynamical systems based on machine learning with high accuracy and interpretability. The proposed method enhances the estimation accuracy for dynamical systems using sparse modeling by incorporating physical constraints derived from Hamiltonian mechanics. Unlike conventional approaches used for estimating governing equations for dynamical systems, we employ a sparse representation of Hamiltonian, allowing for the estimation. Using noisy observational data, the proposed method demonstrates a capability to achieve accurate parameter estimation and extraction of essential nonlinear terms. In addition, it is shown that estimations based on energy conservation principles exhibit superior accuracy in long-term predictions. These results collectively indicate that the proposed method accurately estimates dynamical systems while maintaining interpretability.">
@@ -593,7 +593,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2021-06-09</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2/index.html b/site/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2/index.html
index 5958c10c..10178080 100644
--- a/site/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2/index.html
+++ b/site/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:56 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:42 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -496,7 +496,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-11-20</td>
           <td>Physical review. E</td>
-          <td>9</td>
+          <td>10</td>
           <td>51</td>
         </tr>
 
@@ -508,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-01-13</td>
           <td>Proceedings of the Royal Society A</td>
-          <td>70</td>
-          <td>65</td>
+          <td>75</td>
+          <td>67</td>
         </tr>
 
         <tr id="Making accurate forecasts for a complex system is a challenge in various practical applications. The major difficulty in solving such a problem concerns nonlinear spatiotemporal dynamics with time-varying characteristics. Takens' delay embedding theory provides a way to transform high-dimensional spatial information into temporal information. In this work, by combining delay embedding theory and deep learning techniques, we propose a novel framework, delay-embedding-based forecast Machine (DEFM), to predict the future values of a target variable in a self-supervised and multistep-ahead manner based on high-dimensional observations. With a three-module spatiotemporal architecture, the DEFM leverages deep neural networks to effectively extract both the spatially and temporally associated information from the observed time series even with time-varying parameters or additive noise. The DEFM can accurately predict future information by transforming spatiotemporal information to the delay embeddings of a target variable. The efficacy and precision of the DEFM are substantiated through applications in three spatiotemporally chaotic systems: a 90-dimensional (90D) coupled Lorenz system, the Lorenz 96 system, and the Kuramoto-Sivashinsky equation with inhomogeneity. Additionally, the performance of the DEFM is evaluated on six real-world datasets spanning various fields. Comparative experiments with five prediction methods illustrate the superiority and robustness of the DEFM and show the great potential of the DEFM in temporal information mining and forecasting.">
@@ -521,7 +521,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2020-05-16</td>
           <td>Chaos</td>
           <td>1</td>
-          <td>80</td>
+          <td>83</td>
         </tr>
 
         <tr id="Making accurate forecasts for a complex system is a challenge in various practical applications. The major difficulty in solving such a problem concerns nonlinear spatiotemporal dynamics with time-varying characteristics. Takens' delay embedding theory provides a way to transform high-dimensional spatial information into temporal information. In this work, by combining delay embedding theory and deep learning techniques, we propose a novel framework, delay-embedding-based forecast Machine (DEFM), to predict the future values of a target variable in a self-supervised and multistep-ahead manner based on high-dimensional observations. With a three-module spatiotemporal architecture, the DEFM leverages deep neural networks to effectively extract both the spatially and temporally associated information from the observed time series even with time-varying parameters or additive noise. The DEFM can accurately predict future information by transforming spatiotemporal information to the delay embeddings of a target variable. The efficacy and precision of the DEFM are substantiated through applications in three spatiotemporally chaotic systems: a 90-dimensional (90D) coupled Lorenz system, the Lorenz 96 system, and the Kuramoto-Sivashinsky equation with inhomogeneity. Additionally, the performance of the DEFM is evaluated on six real-world datasets spanning various fields. Comparative experiments with five prediction methods illustrate the superiority and robustness of the DEFM and show the great potential of the DEFM in temporal information mining and forecasting.">
@@ -533,7 +533,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2020-05-16</td>
           <td>Chaos</td>
           <td>1</td>
-          <td>80</td>
+          <td>83</td>
         </tr>
 
         <tr id="Many complex phenomena, from weather systems to heartbeat rhythm patterns, are effectively modeled as low-dimensional dynamical systems. Such systems may behave chaotically under certain conditions, and so the ability to detect chaos based on empirical measurement is an important step in characterizing and predicting these processes. Classifying a system as chaotic usually requires estimating its largest Lyapunov exponent, which quantifies the average rate of convergence or divergence of initially close trajectories in state space, and for which a positive value is generally accepted as an operational definition of chaos. Estimating the largest Lyapunov exponent from observations of a process is especially challenging in systems affected by dynamical noise, which is the case for many models of real-world processes, in particular models of biological systems. We describe a novel method for estimating the largest Lyapunov exponent from data, based on training Deep Learning models on synthetically generated trajectories, and demonstrate that this method yields accurate and noise-robust predictions given relatively short inputs and across a range of different dynamical systems. Our method is unique in that it can analyze tree-shaped data, a ubiquitous topology in biological settings, and specifically in dynamics over lineages of cells or organisms. We also characterize the types of input information extracted by our models for their predictions, allowing for a deeper understanding into the different ways by which chaos can be analyzed in different topologies.">
@@ -556,8 +556,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-06-01</td>
           <td>Science Advances</td>
-          <td>44</td>
-          <td>49</td>
+          <td>50</td>
+          <td>50</td>
         </tr>
 
         <tr id="We propose a universal method for data-driven modeling of complex nonlinear dynamics from time-resolved snapshot data without prior knowledge. Complex nonlinear dynamics govern many fields of science and engineering. Data-driven dynamic modeling often assumes a low-dimensional subspace or manifold for the state. We liberate ourselves from this assumption by proposing cluster-based network modeling (CNM) bridging machine learning, network science, and statistical physics. CNM only assumes smoothness of the dynamics in the state space, robustly describes short- and long-term behavior and is fully automatable as it does not rely on application-specific knowledge. CNM is demonstrated for the Lorenz attractor, ECG heartbeat signals, Kolmogorov flow, and a high-dimensional actuated turbulent boundary layer. Even the notoriously difficult modeling benchmark of rare events in the Kolmogorov flow is solved. This automatable universal data-driven representation of complex nonlinear dynamics complements and expands network connectivity science and promises new fast-track avenues to understand, estimate, predict and control complex systems in all scientific fields.">
@@ -569,7 +569,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2020-10-30</td>
           <td>arXiv: Data Analysis, Statistics and Probability</td>
           <td>1</td>
-          <td>49</td>
+          <td>50</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba/index.html b/site/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba/index.html
index 9130492e..34889c39 100644
--- a/site/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba/index.html
+++ b/site/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:09 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:08 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-06-26</td>
           <td>DBLP</td>
-          <td>9</td>
-          <td>16</td>
+          <td>12</td>
+          <td>17</td>
         </tr>
 
         <tr id="Irregularly-sampled time series (ITS) are native to high-impact domains like healthcare, where measurements are collected over time at uneven intervals. However, for many classification problems, only small portions of long time series are often relevant to the class label. In this case, existing ITS models often fail to classify long series since they rely on careful imputation, which easily over- or under-samples the relevant regions. Using this insight, we then propose CAT, a model that classifies multivariate ITS by explicitly seeking highly-relevant portions of an input series' timeline. CAT achieves this by integrating three components: (1) A Moment Network learns to seek relevant moments in an ITS's continuous timeline using reinforcement learning. (2) A Receptor Network models the temporal dynamics of both observations and their timing localized around predicted moments. (3) A recurrent Transition Model models the sequence of transitions between these moments, cultivating a representation with which the series is classified. Using synthetic and real data, we find that CAT outperforms ten state-of-the-art methods by finding short signals in long irregular time series.">
@@ -497,7 +497,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2023-02-08</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>38</td>
+          <td>39</td>
         </tr>
 
         <tr id="Prevalent in many real-world settings such as healthcare, irregular time series are challenging to formulate predictions from. It is difficult to infer the value of a feature at any given time when observations are sporadic, as it could take on a range of values depending on when it was last observed. To characterize this uncertainty we present EDICT, a strategy that learns an evidential distribution over irregular time series in continuous time. This distribution enables well-calibrated and flexible inference of partially observed features at any time of interest, while expanding uncertainty temporally for sparse, irregular observations. We demonstrate that EDICT attains competitive performance on challenging time series classification tasks and enabling uncertainty-guided inference when encountering noisy data.">
@@ -520,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-03-01</td>
           <td>ArXiv</td>
-          <td>12</td>
+          <td>17</td>
           <td>65</td>
         </tr>
 
@@ -532,10 +532,22 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-03-01</td>
           <td>ArXiv</td>
-          <td>12</td>
+          <td>17</td>
           <td>65</td>
         </tr>
 
+        <tr id="Irregular time series, where data points are recorded at uneven intervals, are prevalent in healthcare settings, such as emergency wards where vital signs and laboratory results are captured at varying times. This variability, which reflects critical fluctuations in patient health, is essential for informed clinical decision-making. Existing self-supervised learning research on irregular time series often relies on generic pretext tasks like forecasting, which may not fully utilise the signal provided by irregular time series. There is a significant need for specialised pretext tasks designed for the characteristics of irregular time series to enhance model performance and robustness, especially in scenarios with limited data availability. This paper proposes a novel pretraining framework, EMIT, an event-based masking for irregular time series. EMIT focuses on masking-based reconstruction in the latent space, selecting masking points based on the rate of change in the data. This method preserves the natural variability and timing of measurements while enhancing the model's ability to process irregular intervals without losing essential information. Extensive experiments on the MIMIC-III and PhysioNet Challenge datasets demonstrate the superior performance of our event-based masking strategy. The code has been released at https://github.com/hrishi-ds/EMIT.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/5e13c5354481232bed4c0fb1e4b6a37d07f32e52" target='_blank'>EMIT- Event-Based Masked Auto Encoding for Irregular Time Series</a></td>
+          <td>
+            Hrishikesh Patel, Ruihong Qiu, Adam Irwin, S. Sadiq, Sen Wang
+          </td>
+          <td>2024-09-25</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>2</td>
+        </tr>
+
         <tr id="To analyze multivariate time series, most previous methods assume regular subsampling of time series, where the interval between adjacent measurements and the number of samples remain unchanged. Practically, data collection systems could produce irregularly sampled time series due to sensor failures and interventions. However, existing methods designed for regularly sampled multivariate time series cannot directly handle irregularity owing to misalignment along both temporal and variate dimensions. To fill this gap, we propose Compatible Transformer (CoFormer), a transformer-based encoder to achieve comprehensive temporal-interaction feature learning for each individual sample in irregular multivariate time series. In CoFormer, we view each sample as a unique variate-time point and leverage intra-variate/inter-variate attentions to learn sample-wise temporal/interaction features based on intra-variate/inter-variate neighbors. With CoFormer as the core, we can analyze irregularly sampled multivariate time series for many downstream tasks, including classification and prediction. We conduct extensive experiments on 3 real-world datasets and validate that the proposed CoFormer significantly and consistently outperforms existing methods. Code will be avilable at https://github.com/MediaBrain-SJTU/CoFormer.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/f6f708484a85b2b96d6f6764724b950cab167bb7" target='_blank'>Compatible Transformer for Irregularly Sampled Multivariate Time Series</a></td>
@@ -555,9 +567,9 @@ <h3 data-intro='Recommendations for the article'>
             Steven Cheng-Xian Li, Benjamin M Marlin
           </td>
           <td>2020-07-12</td>
-          <td>ArXiv, DBLP, MAG</td>
-          <td>48</td>
-          <td>33</td>
+          <td>ArXiv, MAG, DBLP</td>
+          <td>54</td>
+          <td>34</td>
         </tr>
 
         <tr id="We consider a sequence of related multivariate time series learning tasks, such as predicting failures for different instances of a machine from time series of multi-sensor data, or activity recognition tasks over different individuals from multiple wearable sensors. We focus on two under-explored practical challenges arising in such settings: (i) Each task may have a different subset of sensors, i.e., providing different partial observations of the underlying ‘system’. This restriction can be due to different manufacturers in the former case, and people wearing more or less measurement devices in the latter (ii) We are not allowed to store or re-access data from a task once it has been observed at the task level. This may be due to privacy considerations in the case of people, or legal restrictions placed by machine owners. Nevertheless, we would like to (a) improve performance on subsequent tasks using experience from completed tasks as well as (b) continue to perform better on past tasks, e.g., update the model and improve predictions on even the first machine after learning from subsequently observed ones. We note that existing continual learning methods do not take into account variability in input dimensions arising due to different subsets of sensors being available across tasks, and struggle to adapt to such variable input dimensions (VID) tasks. In this work, we address this shortcoming of existing methods. To this end, we learn task-specific generative models and classifiers, and use these to augment data for target tasks. Since the input dimensions across tasks vary, we propose a novel conditioning module based on graph neural networks to aid a standard recurrent neural network. We evaluate the efficacy of the proposed approach on three publicly available datasets corresponding to two activity recognition tasks (classification) and one prognostics task (regression). We demonstrate that it is possible to significantly enhance the performance on future and previous tasks while learning continuously from VID tasks without storing data.">
@@ -568,20 +580,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-12-01</td>
           <td>2021 IEEE International Conference on Data Mining (ICDM)</td>
-          <td>16</td>
-          <td>27</td>
-        </tr>
-
-        <tr id="Multivariate time series (MTS) data are becoming increasingly ubiquitous in diverse domains, e.g., IoT systems, health informatics, and 5G networks. To obtain an effective representation of MTS data, it is not only essential to consider unpredictable dynamics and highly variable lengths of these data but also important to address the irregularities in the sampling rates of MTS. Existing parametric approaches rely on manual hyperparameter tuning and may cost a huge amount of labor effort. Therefore, it is desirable to learn the representation automatically and efficiently. To this end, we propose an autonomous representation learning approach for multivariate time series (TimeAutoML) with irregular sampling rates and variable lengths. As opposed to previous works, we first present a representation learning pipeline in which the configuration and hyperparameter optimization are fully automatic and can be tailored for various tasks, e.g., anomaly detection, clustering, etc. Next, a negative sample generation approach and an auxiliary classification task are developed and integrated within TimeAutoML to enhance its representation capability. Extensive empirical studies on real-world datasets demonstrate that the proposed TimeAutoML outperforms competing approaches on various tasks by a large margin. In fact, it achieves the best anomaly detection performance among all comparison algorithms on 78 out of all 85 UCR datasets, acquiring up to 20% performance improvement in terms of AUC score.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/60773c01dd3c76be0e1d3bc429cf474e5f9d8519" target='_blank'>TimeAutoML: Autonomous Representation Learning for Multivariate Irregularly Sampled Time Series</a></td>
-          <td>
-            Yang Jiao, Kai Yang, Shaoyu Dou, Pan Luo, Sijia Liu, Dongjin Song
-          </td>
-          <td>2020-10-04</td>
-          <td>ArXiv</td>
-          <td>6</td>
-          <td>22</td>
+          <td>17</td>
+          <td>28</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01/index.html b/site/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01/index.html
index 4fa95794..a13c9df8 100644
--- a/site/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01/index.html
+++ b/site/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:03 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:46 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-09-04</td>
           <td>Scientific Reports</td>
-          <td>2</td>
-          <td>23</td>
+          <td>3</td>
+          <td>24</td>
         </tr>
 
         <tr id="Modern societies have an abundance of data yet good system models are rare. Unfortunately, many of the current system identification and machine learning techniques fail to generalize outside of the training set, producing models that violate basic physical laws. This work proposes a novel method for the Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI). SINDy-SI is an iterative method that uses Sum-of-Squares (SOS) programming to learn optimally fitted models while guaranteeing that the learned model satisfies side information, such as symmetries and physical laws. Guided by the principle of Occam's razor, that the simplest or most regularized best fitted model is typically the superior choice, during each iteration SINDy-SI prunes the basis functions associated with small coefficients, yielding a sparse dynamical model upon termination. Through several numerical experiments we will show how the combination of side information constraints and sparse polynomial representation cultivates dynamical models that obey known physical laws while displaying impressive generalized performance beyond the training set.">
@@ -508,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>111</td>
-          <td>65</td>
+          <td>119</td>
+          <td>67</td>
         </tr>
 
         <tr id="Data-driven modeling frameworks that adopt sparse regression techniques, such as sparse identification of nonlinear dynamics (SINDy) and its modifications, are developed to resolve difficulties in extracting underlying dynamics from experimental data. In contrast to neural-network-based methods, these methods are designed to obtain white-box analytical models. In this work, we incorporate the concept of SINDy and knowledge in the field of classical mechanics to identify interpretable and sparse expressions of total energy and the Lagrangian that shelters the hidden dynamics. Moreover, our method (hereafter referred as Lagrangian-SINDy) is developed to use knowledge of simple systems that form the system being analyzed to ensure the likelihood of correct results and to improve the learning pace. Lagrangian-SINDy is highly accurate in discovering interpretable dynamics via energy-related physical quantities. Its performance is validated with three popular multi-DOF nonlinear dynamical systems, namely the spherical pendulum, double pendulum and cart-pendulum system. Comparisons with other SINDy-based methods are made and Lagrangian-SINDy is found to provide the most compact analytical models.">
@@ -520,8 +520,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-01-31</td>
           <td>IEEE Robotics and Automation Letters</td>
-          <td>23</td>
-          <td>23</td>
+          <td>24</td>
+          <td>24</td>
         </tr>
 
         <tr id="None">
@@ -532,7 +532,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
 
@@ -544,7 +544,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-11-10</td>
           <td>SSRN Electronic Journal</td>
-          <td>10</td>
+          <td>14</td>
           <td>33</td>
         </tr>
 
@@ -552,12 +552,24 @@ <h3 data-intro='Recommendations for the article'>
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/7a3166991a8c4161d83c4cdb27033c02f3135cc0" target='_blank'>A Robust SINDy Approach by Combining Neural Networks and an Integral Form</a></td>
           <td>
-            A. Forootani, P. Goyal, P. Benner
+            Ali Forootani, P. Goyal, P. Benner
           </td>
           <td>2023-09-13</td>
           <td>ArXiv</td>
           <td>2</td>
-          <td>14</td>
+          <td>15</td>
+        </tr>
+
+        <tr id="Identifying dynamical systems characterized by nonlinear parameters presents significant challenges in deriving mathematical models that enhance understanding of physics. Traditional methods, such as Sparse Identification of Nonlinear Dynamics (SINDy) and symbolic regression, can extract governing equations from observational data; however, they also come with distinct advantages and disadvantages. This paper introduces a novel method within the SINDy framework, termed ADAM-SINDy, which synthesizes the strengths of established approaches by employing the ADAM optimization algorithm. This facilitates the simultaneous optimization of nonlinear parameters and coefficients associated with nonlinear candidate functions, enabling precise parameter estimation without requiring prior knowledge of nonlinear characteristics such as trigonometric frequencies, exponential bandwidths, or polynomial exponents, thereby addressing a key limitation of SINDy. Through an integrated global optimization, ADAM-SINDy dynamically adjusts all unknown variables in response to data, resulting in an adaptive identification procedure that reduces the sensitivity to the library of candidate functions. The performance of the ADAM-SINDy methodology is demonstrated across a spectrum of dynamical systems, including benchmark coupled nonlinear ordinary differential equations such as oscillators, chaotic fluid flows, reaction kinetics, pharmacokinetics, as well as nonlinear partial differential equations (wildfire transport). The results demonstrate significant improvements in identifying parameterized dynamical systems and underscore the importance of concurrently optimizing all parameters, particularly those characterized by nonlinear parameters. These findings highlight the potential of ADAM-SINDy to extend the applicability of the SINDy framework in addressing more complex challenges in dynamical system identification.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/0c327ed253a2c7b658940860bcf0a9e077d634b6" target='_blank'>ADAM-SINDy: An Efficient Optimization Framework for Parameterized Nonlinear Dynamical System Identification</a></td>
+          <td>
+            Siva Viknesh, Younes Tatari, Amirhossein Arzani
+          </td>
+          <td>2024-10-21</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>24</td>
         </tr>
 
         <tr id="Symbolic Regression (SR) is a widely studied field of research that aims to infer symbolic expressions from data. A popular approach for SR is the Sparse Identification of Nonlinear Dynamical Systems (\sindy) framework, which uses sparse regression to identify governing equations from data. This study introduces an enhanced method, Nested SINDy, that aims to increase the expressivity of the SINDy approach thanks to a nested structure. Indeed, traditional symbolic regression and system identification methods often fail with complex systems that cannot be easily described analytically. Nested SINDy builds on the SINDy framework by introducing additional layers before and after the core SINDy layer. This allows the method to identify symbolic representations for a wider range of systems, including those with compositions and products of functions. We demonstrate the ability of the Nested SINDy approach to accurately find symbolic expressions for simple systems, such as basic trigonometric functions, and sparse (false but accurate) analytical representations for more complex systems. Our results highlight Nested SINDy's potential as a tool for symbolic regression, surpassing the traditional SINDy approach in terms of expressivity. However, we also note the challenges in the optimization process for Nested SINDy and suggest future research directions, including the designing of a more robust methodology for the optimization process. This study proves that Nested SINDy can effectively discover symbolic representations of dynamical systems from data, offering new opportunities for understanding complex systems through data-driven methods.">
@@ -581,19 +593,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2024-03-01</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>1</td>
-        </tr>
-
-        <tr id="None">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/7be51140c9c66207fc1baa0ca92a691726cbd900" target='_blank'>Weak SINDy For Partial Differential Equations</a></td>
-          <td>
-            D. Messenger, D. Bortz
-          </td>
-          <td>2020-07-06</td>
-          <td>Journal of computational physics</td>
-          <td>126</td>
-          <td>20</td>
+          <td>2</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/4a111f7a3b56d0468f13104999844885157ef17d/index.html b/site/recommendations/4a111f7a3b56d0468f13104999844885157ef17d/index.html
index 68518a4b..227bb1fb 100644
--- a/site/recommendations/4a111f7a3b56d0468f13104999844885157ef17d/index.html
+++ b/site/recommendations/4a111f7a3b56d0468f13104999844885157ef17d/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:24 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:17 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -476,15 +476,51 @@ <h3 data-intro='Recommendations for the article'>
   </thead>
   <tbody>
 
+        <tr id="Time series foundation models excel in zero-shot forecasting, handling diverse tasks without explicit training. However, the advancement of these models has been hindered by the lack of comprehensive benchmarks. To address this gap, we introduce the General Time Series Forecasting Model Evaluation, GIFT-Eval, a pioneering benchmark aimed at promoting evaluation across diverse datasets. GIFT-Eval encompasses 28 datasets over 144,000 time series and 177 million data points, spanning seven domains, 10 frequencies, multivariate inputs, and prediction lengths ranging from short to long-term forecasts. To facilitate the effective pretraining and evaluation of foundation models, we also provide a non-leaking pretraining dataset containing approximately 230 billion data points. Additionally, we provide a comprehensive analysis of 17 baselines, which includes statistical models, deep learning models, and foundation models. We discuss each model in the context of various benchmark characteristics and offer a qualitative analysis that spans both deep learning and foundation models. We believe the insights from this analysis, along with access to this new standard zero-shot time series forecasting benchmark, will guide future developments in time series foundation models. The codebase, datasets, and a leaderboard showing all the results in detail will be available soon.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/8bf78baccb9e56ffef46080b9b34752f23d1911c" target='_blank'>GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation</a></td>
+          <td>
+            Taha İbrahim Aksu, Gerald Woo, Juncheng Liu, Xu Liu, Chenghao Liu, Silvio Savarese, Caiming Xiong, Doyen Sahoo
+          </td>
+          <td>2024-10-14</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>25</td>
+        </tr>
+
+        <tr id="Time Series Forecasting (TSF) is key functionality in numerous fields, including in finance, weather services, and energy management. While TSF methods are emerging these days, many of them require domain-specific data collection and model training and struggle with poor generalization performance on new domains. Foundation models aim to overcome this limitation. Pre-trained on large-scale language or time series data, they exhibit promising inferencing capabilities in new or unseen data. This has spurred a surge in new TSF foundation models. We propose a new benchmark, FoundTS, to enable thorough and fair evaluation and comparison of such models. FoundTS covers a variety of TSF foundation models, including those based on large language models and those pretrained on time series. Next, FoundTS supports different forecasting strategies, including zero-shot, few-shot, and full-shot, thereby facilitating more thorough evaluations. Finally, FoundTS offers a pipeline that standardizes evaluation processes such as dataset splitting, loading, normalization, and few-shot sampling, thereby facilitating fair evaluations. Building on this, we report on an extensive evaluation of TSF foundation models on a broad range of datasets from diverse domains and with different statistical characteristics. Specifically, we identify pros and cons and inherent limitations of existing foundation models, and we identify directions for future model design. We make our code and datasets available at https://anonymous.4open.science/r/FoundTS-C2B0.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/c245a7d561855d7bef6c758e45e772129b52dacb" target='_blank'>FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting</a></td>
+          <td>
+            Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang
+          </td>
+          <td>2024-10-15</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>27</td>
+        </tr>
+
+        <tr id="Deep learning for time series forecasting has seen significant advancements over the past decades. However, despite the success of large-scale pre-training in language and vision domains, pre-trained time series models remain limited in scale and operate at a high cost, hindering the development of larger capable forecasting models in real-world applications. In response, we introduce Time-MoE, a scalable and unified architecture designed to pre-train larger, more capable forecasting foundation models while reducing inference costs. By leveraging a sparse mixture-of-experts (MoE) design, Time-MoE enhances computational efficiency by activating only a subset of networks for each prediction, reducing computational load while maintaining high model capacity. This allows Time-MoE to scale effectively without a corresponding increase in inference costs. Time-MoE comprises a family of decoder-only transformer models that operate in an auto-regressive manner and support flexible forecasting horizons with varying input context lengths. We pre-trained these models on our newly introduced large-scale data Time-300B, which spans over 9 domains and encompassing over 300 billion time points. For the first time, we scaled a time series foundation model up to 2.4 billion parameters, achieving significantly improved forecasting precision. Our results validate the applicability of scaling laws for training tokens and model size in the context of time series forecasting. Compared to dense models with the same number of activated parameters or equivalent computation budgets, our models consistently outperform them by large margin. These advancements position Time-MoE as a state-of-the-art solution for tackling real-world time series forecasting challenges with superior capability, efficiency, and flexibility.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/2cb3044ef42c7ee022a988864028b80ce977072c" target='_blank'>Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts</a></td>
+          <td>
+            X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin
+          </td>
+          <td>2024-09-24</td>
+          <td>ArXiv</td>
+          <td>2</td>
+          <td>9</td>
+        </tr>
+
         <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4fb78450650894091b0a55b5504a8b0a6f3dec37" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
           <td>
             Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>28</td>
+          <td>69</td>
           <td>18</td>
         </tr>
 
@@ -496,8 +532,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>4</td>
-          <td>66</td>
+          <td>21</td>
+          <td>67</td>
         </tr>
 
         <tr id="We propose an efficient design of Transformer-based models for multivariate time series forecasting and self-supervised representation learning. It is based on two key components: (i) segmentation of time series into subseries-level patches which are served as input tokens to Transformer; (ii) channel-independence where each channel contains a single univariate time series that shares the same embedding and Transformer weights across all the series. Patching design naturally has three-fold benefit: local semantic information is retained in the embedding; computation and memory usage of the attention maps are quadratically reduced given the same look-back window; and the model can attend longer history. Our channel-independent patch time series Transformer (PatchTST) can improve the long-term forecasting accuracy significantly when compared with that of SOTA Transformer-based models. We also apply our model to self-supervised pre-training tasks and attain excellent fine-tuning performance, which outperforms supervised training on large datasets. Transferring of masked pre-trained representation on one dataset to others also produces SOTA forecasting accuracy. Code is available at: https://github.com/yuqinie98/PatchTST.">
@@ -508,56 +544,32 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-11-27</td>
           <td>ArXiv</td>
-          <td>534</td>
-          <td>34</td>
+          <td>724</td>
+          <td>35</td>
         </tr>
 
-        <tr id="Time series forecasting is a critical and challenging task in practical application. Recent advancements in pre-trained foundation models for time series forecasting have gained significant interest. However, current methods often overlook the multi-scale nature of time series, which is essential for accurate forecasting. To address this, we propose HiMTM, a hierarchical multi-scale masked time series modeling with self-distillation for long-term forecasting. HiMTM integrates four key components: (1) hierarchical multi-scale transformer (HMT) to capture temporal information at different scales; (2) decoupled encoder-decoder (DED) that directs the encoder towards feature extraction while the decoder focuses on pretext tasks; (3) hierarchical self-distillation (HSD) for multi-stage feature-level supervision signals during pre-training; and (4) cross-scale attention fine-tuning (CSA-FT) to capture dependencies between different scales for downstream tasks. These components collectively enhance multi-scale feature extraction in masked time series modeling, improving forecasting accuracy. Extensive experiments on seven mainstream datasets show that HiMTM surpasses state-of-the-art self-supervised and end-to-end learning methods by a considerable margin of 3.16-68.54\%. Additionally, HiMTM outperforms the latest robust self-supervised learning method, PatchTST, in cross-domain forecasting by a significant margin of 2.3\%. The effectiveness of HiMTM is further demonstrated through its application in natural gas demand forecasting.">
+        <tr id="Developing a foundation model for time series forecasting across diverse domains has attracted significant attention in recent years. Existing works typically assume regularly sampled, well-structured data, limiting their applicability to more generalized scenarios where time series often contain missing values, unequal sequence lengths, and irregular time intervals between measurements. To cover diverse domains and handle variable regularities, we propose FlexTSF, a universal time series forecasting model that possesses better generalization and natively support both regular and irregular time series. FlexTSF produces forecasts in an autoregressive manner and incorporates three novel designs: VT-Norm, a normalization strategy to ablate data domain barriers, IVP Patcher, a patching module to learn representations from flexibly structured time series, and LED attention, an attention mechanism to seamlessly integrate these two and propagate forecasts with awareness of domain and time information. Experiments on 12 datasets show that FlexTSF outperforms state-of-the-art forecasting models respectively designed for regular and irregular time series. Furthermore, after self-supervised pre-training, FlexTSF shows exceptional performance in both zero-shot and few-show settings for time series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/d945eae22457648e191742c37bcad88a114ce703" target='_blank'>HiMTM: Hierarchical Multi-Scale Masked Time Series Modeling with Self-Distillation for Long-Term Forecasting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/45521591fd8c46068f0a8712b43ce6cdaac1edea" target='_blank'>FlexTSF: A Universal Forecasting Model for Time Series with Variable Regularities</a></td>
           <td>
-            Shubao Zhao, Ming Jin, Zhaoxiang Hou, Che-Sheng Yang, Zengxiang Li, Qingsong Wen, Yi Wang
+            Jingge Xiao, Yile Chen, Gao Cong, Wolfgang Nejdl, Simon Gottschalk
           </td>
-          <td>2024-01-10</td>
+          <td>2024-10-30</td>
           <td>ArXiv</td>
           <td>0</td>
           <td>6</td>
         </tr>
 
-        <tr id="The Transformer is a highly successful deep learning model that has revolutionised the world of artificial neural networks, first in natural language processing and later in computer vision. This model is based on the attention mechanism and is able to capture complex semantic relationships between a variety of patterns present in the input data. Precisely because of these characteristics, the Transformer has recently been exploited for time series forecasting problems, assuming a natural adaptability to the domain of continuous numerical series. Despite the acclaimed results in the literature, some works have raised doubts about the robustness and effectiveness of this approach. In this paper, we further investigate the effectiveness of Transformer-based models applied to the domain of time series forecasting, demonstrate their limitations, and propose a set of alternative models that are better performing and significantly less complex. In particular, we empirically show how simplifying Transformer-based forecasting models almost always leads to an improvement, reaching state of the art performance. We also propose shallow models without the attention mechanism, which compete with the overall state of the art in long time series forecasting, and demonstrate their ability to accurately predict time series over extremely long windows. From a methodological perspective, we show how it is always necessary to use a simple baseline to verify the effectiveness of proposed models, and finally, we conclude the paper with a reflection on recent research paths and the opportunity to follow trends and hypes even where it may not be necessary.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/b98bc7d48357932537fc9ba6035e3b6bd1018ae6" target='_blank'>Two Steps Forward and One Behind: Rethinking Time Series Forecasting with Deep Learning</a></td>
-          <td>
-            Riccardo Ughi, Eugenio Lomurno, Matteo Matteucci
-          </td>
-          <td>2023-04-10</td>
-          <td>ArXiv, DBLP</td>
-          <td>1</td>
-          <td>6</td>
-        </tr>
-
-        <tr id="Time series has been left behind in the era of pre-training and transfer learning. While research in the fields of natural language processing and computer vision are enjoying progressively larger datasets to train massive models, the most popular time series datasets consist of only tens of thousands of time steps, limiting our ability to study the effectiveness of pre-training and scaling. Recent studies have also cast doubt on the need for expressive models and scale. To alleviate these issues, we introduce three large-scale time series forecasting datasets from the cloud operations (CloudOps) domain, the largest having billions of observations, enabling further study into pre-training and scaling of time series models. We build the empirical groundwork for studying pre-training and scaling of time series models and pave the way for future research by identifying a promising candidate architecture. We show that it is a strong zero-shot baseline and benefits from further scaling, both in model and dataset size. Accompanying these datasets and results is a suite of comprehensive benchmark results comparing classical and deep learning baselines to our pre-trained method - achieving a 27% reduction in error on the largest dataset. Code and datasets can be found https://github.com/SalesforceAIResearch/pretrain-time-series-cloudops.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/dbd6463cfff26861f6f3ea8e8d19eff4e6331904" target='_blank'>Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain</a></td>
-          <td>
-            Gerald Woo, Chenghao Liu, Akshat Kumar, Doyen Sahoo
-          </td>
-          <td>2023-10-08</td>
-          <td>ArXiv</td>
-          <td>7</td>
-          <td>22</td>
-        </tr>
-
-        <tr id="Time Series Forecasting plays a crucial role in various fields such as industrial equipment maintenance, meteorology, energy consumption, traffic flow and financial investment. However, despite their considerable advantages over traditional statistical approaches, current deep learning-based predictive models often exhibit a significant deviation between their forecasting outcomes and the ground truth. This discrepancy is largely due to an insufficient emphasis on extracting the sequence's latent information, particularly its global information within the frequency domain and the relationship between different variables. To address this issue, we propose a novel model Frequency-domain Attention In Two Horizons, which decomposes time series into trend and seasonal components using a multi-scale sequence adaptive decomposition and fusion architecture, and processes them separately. FAITH utilizes Frequency Channel feature Extraction Module and Frequency Temporal feature Extraction Module to capture inter-channel relationships and temporal global information in the sequence, significantly improving its ability to handle long-term dependencies and complex patterns. Furthermore, FAITH achieves theoretically linear complexity by modifying the time-frequency domain transformation method, effectively reducing computational costs. Extensive experiments on 6 benchmarks for long-term forecasting and 3 benchmarks for short-term forecasting demonstrate that FAITH outperforms existing models in many fields, such as electricity, weather and traffic, proving its effectiveness and superiority both in long-term and short-term time series forecasting tasks. Our codes and data are available at https://github.com/LRQ577/FAITH.">
+        <tr id="Time series forecasting is a critical and challenging task in practical application. Recent advancements in pre-trained foundation models for time series forecasting have gained significant interest. However, current methods often overlook the multi-scale nature of time series, which is essential for accurate forecasting. To address this, we propose HiMTM, a hierarchical multi-scale masked time series modeling with self-distillation for long-term forecasting. HiMTM integrates four key components: (1) hierarchical multi-scale transformer (HMT) to capture temporal information at different scales; (2) decoupled encoder-decoder (DED) that directs the encoder towards feature extraction while the decoder focuses on pretext tasks; (3) hierarchical self-distillation (HSD) for multi-stage feature-level supervision signals during pre-training; and (4) cross-scale attention fine-tuning (CSA-FT) to capture dependencies between different scales for downstream tasks. These components collectively enhance multi-scale feature extraction in masked time series modeling, improving forecasting accuracy. Extensive experiments on seven mainstream datasets show that HiMTM surpasses state-of-the-art self-supervised and end-to-end learning methods by a considerable margin of 3.16-68.54\%. Additionally, HiMTM outperforms the latest robust self-supervised learning method, PatchTST, in cross-domain forecasting by a significant margin of 2.3\%. The effectiveness of HiMTM is further demonstrated through its application in natural gas demand forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/ca89b6de0dbe31f523495727dc83f8ab5d2ae1c3" target='_blank'>FAITH: Frequency-domain Attention In Two Horizons for Time Series Forecasting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/d945eae22457648e191742c37bcad88a114ce703" target='_blank'>HiMTM: Hierarchical Multi-Scale Masked Time Series Modeling with Self-Distillation for Long-Term Forecasting</a></td>
           <td>
-            Ruiqi Li, Maowei Jiang, Kai Wang, Kaiduo Feng, Quangao Liu, Yue Sun, Xiufang Zhou
+            Shubao Zhao, Ming Jin, Zhaoxiang Hou, Che-Sheng Yang, Zengxiang Li, Qingsong Wen, Yi Wang
           </td>
-          <td>2024-05-22</td>
-          <td>ArXiv</td>
+          <td>2024-01-10</td>
+          <td>Proceedings of the 33rd ACM International Conference on Information and Knowledge Management</td>
           <td>0</td>
-          <td>2</td>
+          <td>7</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14/index.html b/site/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14/index.html
index 26b1fdaf..822bd49a 100644
--- a/site/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14/index.html
+++ b/site/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:32 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:23 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -497,7 +497,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2024-08-28</td>
           <td>IEEE transactions on neural networks and learning systems</td>
           <td>0</td>
-          <td>14</td>
+          <td>15</td>
         </tr>
 
         <tr id="Unsupervised domain adaptation methods aim at generalizing well on unlabeled test data that may have a different (shifted) distribution from the training data. Such methods are typically developed on image data, and their application to time series data is less explored. Existing works on time series domain adaptation suffer from inconsistencies in evaluation schemes, datasets, and backbone neural network architectures. Moreover, labeled target data are often used for model selection, which violates the fundamental assumption of unsupervised domain adaptation. To address these issues, we develop a benchmarking evaluation suite (AdaTime) to systematically and fairly evaluate different domain adaptation methods on time series data. Specifically, we standardize the backbone neural network architectures and benchmarking datasets, while also exploring more realistic model selection approaches that can work with no labeled data or just a few labeled samples. Our evaluation includes adapting state-of-the-art visual domain adaptation methods to time series data as well as the recent methods specifically developed for time series data. We conduct extensive experiments to evaluate 11 state-of-the-art methods on five representative datasets spanning 50 cross-domain scenarios. Our results suggest that with careful selection of hyper-parameters, visual domain adaptation methods are competitive with methods proposed for time series domain adaptation. In addition, we find that hyper-parameters could be selected based on realistic model selection approaches. Our work unveils practical insights for applying domain adaptation methods on time series data and builds a solid foundation for future works in the field. The code is available at github.com/emadeldeen24/AdaTime.">
@@ -508,8 +508,32 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-03-15</td>
           <td>ACM Transactions on Knowledge Discovery from Data</td>
-          <td>40</td>
-          <td>37</td>
+          <td>51</td>
+          <td>39</td>
+        </tr>
+
+        <tr id="Unsupervised domain adaptation (UDA) of time series aims to teach models to identify consistent patterns across various temporal scenarios, disregarding domain-specific differences, which can maintain their predictive accuracy and effectively adapt to new domains. However, existing UDA methods struggle to adequately extract and align both global and local features in time series data. To address this issue, we propose the Local-Global Representation Alignment framework (LogoRA), which employs a two-branch encoder, comprising a multi-scale convolutional branch and a patching transformer branch. The encoder enables the extraction of both local and global representations from time series. A fusion module is then introduced to integrate these representations, enhancing domain-invariant feature alignment from multi-scale perspectives. To achieve effective alignment, LogoRA employs strategies like invariant feature learning on the source domain, utilizing triplet loss for fine alignment and dynamic time warping-based feature alignment. Additionally, it reduces source-target domain gaps through adversarial training and per-class prototype alignment. Our evaluations on four time-series datasets demonstrate that LogoRA outperforms strong baselines by up to $12.52\%$, showcasing its superiority in time series UDA tasks.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/9e46960ff506775c47d101a6b9c61e32073d772a" target='_blank'>LogoRA: Local-Global Representation Alignment for Robust Time Series Classification</a></td>
+          <td>
+            Huanyu Zhang, Yi-Fan Zhang, Zhang Zhang, Qingsong Wen, Liang Wang
+          </td>
+          <td>2024-09-12</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>4</td>
+        </tr>
+
+        <tr id="Unlike images and natural language tokens, time series data is highly semantically sparse, resulting in labor-intensive label annotations. Unsupervised and Semi-supervised Domain Adaptation (UDA and SSDA) have demonstrated efficiency in addressing this issue by utilizing pre-labeled source data to train on unlabeled or partially labeled target data. However, in domain adaptation methods designed for downstream classification tasks, directly adapting labeled source samples with unlabelled target samples often results in similar distributions across various classes, thereby compromising the performance of the target classification task. To tackle this challenge, we proposed a Global-Local Alignment Domain Adaptation (GLA-DA) method for multivariate time series data. Data from two domains were initially encoded to align in an intermediate feature space adversarially, achieving Global Feature Alignment (GFA). Subsequently, GLA-DA leveraged the consistency between similarity-based and deep learning-based models to assign pseudo labels to unlabeled target data. This process aims to preserve differences among data with distinct labels by aligning the samples with the same class labels together, achieving Local Class Alignment (LCA). We implemented GLA-DA in both UDA and SSDA scenarios, showcasing its superiority over state-of-the-art methods through extensive experiments on various public datasets. Ablation experiments underscored the significance of key components within GLA-DA.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/89d5e78c21b502378b3cfeadc613d4cfe0046b31" target='_blank'>GLA-DA: Global-Local Alignment Domain Adaptation for Multivariate Time Series</a></td>
+          <td>
+            Gang Tu, Dan Li, Bingxin Lin, Zibin Zheng, See-Kiong Ng
+          </td>
+          <td>2024-10-09</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>1</td>
         </tr>
 
         <tr id="Domain generalization methods aim to learn models robust to domain shift with data from a limited number of source domains and without access to target domain samples during training. Popular domain alignment methods for domain generalization seek to extract domain-invariant features by minimizing the discrepancy between feature distributions across all domains, disregarding inter-domain relationships. In this paper, we instead propose a novel representation learning methodology that selectively enforces prediction consistency between source domains estimated to be closely-related. Specifically, we hypothesize that domains share different class-informative representations, so instead of aligning all domains which can cause negative transfer, we only regularize the discrepancy between closely-related domains. We apply our method to time-series classification tasks and conduct comprehensive experiments on three public real-world datasets. Our method significantly improves over the baseline and achieves better or competitive performance in comparison with state-of-the-art methods in terms of both accuracy and model calibration.">
@@ -532,7 +556,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-02-13</td>
           <td>ArXiv, DBLP</td>
-          <td>57</td>
+          <td>63</td>
           <td>65</td>
         </tr>
 
diff --git a/site/recommendations/5d150cec2775f9bc863760448f14104cc8f42368/index.html b/site/recommendations/5d150cec2775f9bc863760448f14104cc8f42368/index.html
index 04113a28..acacbe12 100644
--- a/site/recommendations/5d150cec2775f9bc863760448f14104cc8f42368/index.html
+++ b/site/recommendations/5d150cec2775f9bc863760448f14104cc8f42368/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:54 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:41 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2019-03-29</td>
           <td>Proceedings of the National Academy of Sciences of the United States of America</td>
-          <td>615</td>
-          <td>65</td>
+          <td>648</td>
+          <td>67</td>
         </tr>
 
         <tr id="None">
@@ -496,7 +496,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
 
@@ -520,8 +520,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>250</td>
-          <td>12</td>
+          <td>289</td>
+          <td>13</td>
         </tr>
 
         <tr id="Discovering the partial differential equations underlying spatio-temporal datasets from very limited and highly noisy observations is of paramount interest in many scientific fields. However, it remains an open question to know when model discovery algorithms based on sparse regression can actually recover the underlying physical processes. In this work, we show the design matrices used to infer the equations by sparse regression can violate the irrepresentability condition (IRC) of the Lasso, even when derived from analytical PDE solutions (i.e. without additional noise). Sparse regression techniques which can recover the true underlying model under violated IRC conditions are therefore required, leading to the introduction of the randomised adaptive Lasso. We show once the latter is integrated within the deep learning model discovery framework DeepMod, a wide variety of nonlinear and chaotic canonical PDEs can be recovered: (1) up to $\mathcal{O}(2)$ higher noise-to-sample ratios than state-of-the-art algorithms, (2) with a single set of hyperparameters, which paves the road towards truly automated model discovery.">
@@ -533,7 +533,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2021-06-22</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
 
         <tr id="Harnessing data to discover the underlying governing laws or equations that describe the behavior of complex physical systems can significantly advance our modeling, simulation and understanding of such systems in various science and engineering disciplines. Recent advances in sparse identification show encouraging success in distilling closed-form governing equations from data for a wide range of nonlinear dynamical systems. However, the fundamental bottleneck of this approach lies in the robustness and scalability with respect to data scarcity and noise. This work introduces a novel physics-informed deep learning framework to discover governing partial differential equations (PDEs) from scarce and noisy data for nonlinear spatiotemporal systems. In particular, this approach seamlessly integrates the strengths of deep neural networks for rich representation learning, automatic differentiation and sparse regression to approximate the solution of system variables, compute essential derivatives, as well as identify the key derivative terms and parameters that form the structure and explicit expression of the PDEs. The efficacy and robustness of this method are demonstrated on discovering a variety of PDE systems with different levels of data scarcity and noise. The resulting computational framework shows the potential for closed-form model discovery in practical applications where large and accurate datasets are intractable to capture.">
@@ -545,7 +545,19 @@ <h3 data-intro='Recommendations for the article'>
           <td>2020-05-05</td>
           <td>ArXiv</td>
           <td>19</td>
-          <td>12</td>
+          <td>13</td>
+        </tr>
+
+        <tr id="The combination of machine learning (ML) and sparsity-promoting techniques is enabling direct extraction of governing equations from data, revolutionizing computational modeling in diverse fields of science and engineering. The discovered dynamical models could be used to address challenges in climate science, neuroscience, ecology, finance, epidemiology, and beyond. However, most existing sparse identification methods for discovering dynamical systems treat the whole system as one without considering the interactions between subsystems. As a result, such models are not able to capture small changes in the emergent system behavior. To address this issue, we developed a new method called Sparse Identification of Nonlinear Dynamical Systems from Graph-structured data (SINDyG), which incorporates the network structure into sparse regression to identify model parameters that explain the underlying network dynamics. SINDyG discovers the governing equations of network dynamics while offering improvements in accuracy and model simplicity.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/4da518ae886702aed27abdd8f94fea3042ab9377" target='_blank'>Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems</a></td>
+          <td>
+            Mohammad Amin Basiri, Sina Khanmohammadi
+          </td>
+          <td>2024-09-02</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>3</td>
         </tr>
 
         <tr id="In this work, we blend machine learning and dictionary-based learning with numerical analysis tools to discover differential equations from noisy and sparsely sampled measurement data of time-dependent processes. We use the fact that given a dictionary containing large candidate nonlinear functions, dynamical models can often be described by a few appropriately chosen basis functions. As a result, we obtain parsimonious models that can be better interpreted by practitioners, and potentially generalize better beyond the sampling regime than black-box modelling. In this work, we integrate a numerical integration framework with dictionary learning that yields differential equations without requiring or approximating derivative information at any stage. Hence, it is utterly effective for corrupted and sparsely sampled data. We discuss its extension to governing equations, containing rational nonlinearities that typically appear in biological networks. Moreover, we generalized the method to governing equations subject to parameter variations and externally controlled inputs. We demonstrate the efficiency of the method to discover a number of diverse differential equations using noisy measurements, including a model describing neural dynamics, chaotic Lorenz model, Michaelis–Menten kinetics and a parameterized Hopf normal form.">
@@ -556,7 +568,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-05-11</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>38</td>
+          <td>41</td>
           <td>53</td>
         </tr>
 
@@ -569,7 +581,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2014-12-10</td>
           <td>The European Physical Journal Special Topics</td>
           <td>64</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
 
         <tr id="None">
@@ -581,7 +593,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2014-12-01</td>
           <td>The European Physical Journal Special Topics</td>
           <td>7</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c/index.html b/site/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c/index.html
index 70591eac..fae84787 100644
--- a/site/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c/index.html
+++ b/site/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:54 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:41 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,20 +484,20 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-07-01</td>
           <td>Nature Computational Science</td>
-          <td>69</td>
-          <td>20</td>
+          <td>77</td>
+          <td>21</td>
         </tr>
 
         <tr id="Technological advancements have substantially increased computational power and data availability, enabling the application of powerful machine-learning (ML) techniques across various fields. However, our ability to leverage ML methods for scientific discovery, {\it i.e.} to obtain fundamental and formalized knowledge about natural processes, is still in its infancy. In this review, we explore how the scientific community can increasingly leverage ML techniques to achieve scientific discoveries. We observe that the applicability and opportunity of ML depends strongly on the nature of the problem domain, and whether we have full ({\it e.g.}, turbulence), partial ({\it e.g.}, computational biochemistry), or no ({\it e.g.}, neuroscience) {\it a-priori} knowledge about the governing equations and physical properties of the system. Although challenges remain, principled use of ML is opening up new avenues for fundamental scientific discoveries. Throughout these diverse fields, there is a theme that ML is enabling researchers to embrace complexity in observational data that was previously intractable to classic analysis and numerical investigations.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/f7214e1640b55a726c771c629c873920245e0839" target='_blank'>Opportunities for machine learning in scientific discovery</a></td>
           <td>
-            Ricardo Vinuesa, Jean Rabault, Hossein Azizpour, Stefan Bauer, Bingni W. Brunton, Arne Elofsson, Elias Jarlebring, Hedvig Kjellstrom, Stefano Markidis, David Marlevi, Paola Cinnella, S. Brunton
+            Ricardo Vinuesa, Jean Rabault, H. Azizpour, Stefan Bauer, Bingni W. Brunton, Arne Elofsson, Elias Jarlebring, Hedvig Kjellstrom, Stefano Markidis, David Marlevi, Paola Cinnella, S. Brunton
           </td>
           <td>2024-05-07</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
 
         <tr id="None">
@@ -508,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2019-04-20</td>
           <td>J. Comput. Phys.</td>
-          <td>101</td>
-          <td>34</td>
+          <td>104</td>
+          <td>36</td>
         </tr>
 
         <tr id="None">
@@ -520,8 +520,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-07-22</td>
           <td>Communications Physics</td>
-          <td>17</td>
-          <td>94</td>
+          <td>20</td>
+          <td>95</td>
         </tr>
 
         <tr id="The extraction of models from data (in a sense, the “understanding” of the physical laws giving rise to the data) is a fundamental cognitive as well as scientific challenge. We show a geometric/analytic learning algorithm capable of creating minimal descriptions of parametrically dependent unknown nonlinear dynamical systems. This is accomplished by the data-driven discovery of useful intrinsic-state variables and parameters in terms of which one can empirically model the underlying dynamics. We discuss an informed observation geometry that enables us to formulate models without first principles as well as without closed form equations. (See pp. E7865–E7874.)">
@@ -556,7 +556,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-09-25</td>
           <td>ArXiv</td>
-          <td>88</td>
+          <td>90</td>
           <td>17</td>
         </tr>
 
@@ -568,7 +568,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-01-30</td>
           <td>J. Comput. Phys.</td>
-          <td>32</td>
+          <td>33</td>
           <td>47</td>
         </tr>
 
@@ -581,8 +581,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-01-13</td>
           <td>ArXiv</td>
-          <td>493</td>
-          <td>25</td>
+          <td>509</td>
+          <td>26</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37/index.html b/site/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37/index.html
index 05fcd23c..7fbd134a 100644
--- a/site/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37/index.html
+++ b/site/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:32 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:21 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,7 +484,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-04-23</td>
           <td>Sensors (Basel, Switzerland)</td>
-          <td>22</td>
+          <td>28</td>
           <td>77</td>
         </tr>
 
@@ -509,7 +509,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2023-06-10</td>
           <td>2023 31st European Signal Processing Conference (EUSIPCO)</td>
           <td>0</td>
-          <td>38</td>
+          <td>39</td>
         </tr>
 
         <tr id="Significant progress has been made in representation learning, especially with recent success on self-supervised contrastive learning. However, for time series with less intuitive or semantic meaning, sampling bias may be inevitably encountered in unsupervised approaches. Although supervised contrastive learning has shown superior performance by leveraging label information, it may also suffer from class collapse. In this study, we consider a realistic scenario in industry with limited annotation information available. A supervised contrastive framework is developed for high-frequency time series representation and classification, wherein a novel variant of supervised contrastive loss is proposed to include multiple augmentations while induce spread within each class. Experiments on four mainstream public datasets as well as a series of sensitivity and ablation analyses demonstrate that the learned representations are effective and robust compared with the direct supervised learning and self-supervised learning, notably under the minimal few-shot situation.">
@@ -520,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-06-26</td>
           <td>DBLP</td>
-          <td>2</td>
+          <td>4</td>
           <td>140</td>
         </tr>
 
@@ -532,8 +532,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-10-21</td>
           <td>ArXiv</td>
-          <td>21</td>
-          <td>3</td>
+          <td>30</td>
+          <td>4</td>
         </tr>
 
         <tr id="Unsupervised domain adaptation (UDA) provides a strategy for improving machine learning performance in data-rich (target) domains where ground truth labels are inaccessible but can be found in related (source) domains. In cases where meta-domain information such as label distributions is available, weak supervision can further boost performance. We propose a novel framework, CALDA, to tackle these two problems. CALDA synergistically combines the principles of contrastive learning and adversarial learning to robustly support multi-source UDA (MS-UDA) for time series data. Similar to prior methods, CALDA utilizes adversarial learning to align source and target feature representations. Unlike prior approaches, CALDA additionally leverages cross-source label information across domains. CALDA pulls examples with the same label close to each other, while pushing apart examples with different labels, reshaping the space through contrastive learning. Unlike prior contrastive adaptation methods, CALDA requires neither data augmentation nor pseudo labeling, which may be more challenging for time series. We empirically validate our proposed approach. Based on results from human activity recognition, electromyography, and synthetic datasets, we find utilizing cross-source information improves performance over prior time series and contrastive methods. Weak supervision further improves performance, even in the presence of noise, allowing CALDA to offer generalizable strategies for MS-UDA.">
@@ -544,7 +544,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-09-30</td>
           <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
-          <td>16</td>
+          <td>21</td>
           <td>72</td>
         </tr>
 
@@ -557,7 +557,19 @@ <h3 data-intro='Recommendations for the article'>
           <td>2023-08-04</td>
           <td>Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining</td>
           <td>0</td>
-          <td>24</td>
+          <td>25</td>
+        </tr>
+
+        <tr id="In this work, we investigate the time series representation learning problem using self-supervised techniques. Contrastive learning is well-known in this area as it is a powerful method for extracting information from the series and generating task-appropriate representations. Despite its proficiency in capturing time series characteristics, these techniques often overlook a critical factor - the inherent noise in this type of data, a consideration usually emphasized in general time series analysis. Moreover, there is a notable absence of attention to developing efficient yet lightweight encoder architectures, with an undue focus on delivering contrastive losses. Our work address these gaps by proposing an innovative training strategy that promotes consistent representation learning, accounting for the presence of noise-prone signals in natural time series. Furthermore, we propose an encoder architecture that incorporates dilated convolution within the Inception block, resulting in a scalable and robust network with a wide receptive field. Experimental findings underscore the effectiveness of our method, consistently outperforming state-of-the-art approaches across various tasks, including forecasting, classification, and abnormality detection. Notably, our method attains the top rank in over two-thirds of the classification UCR datasets, utilizing only 40% of the parameters compared to the second-best approach. Our source code for CoInception framework is accessible at https://github.com/anhduy0911/CoInception.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/b1909388ab7a548f6b27dc95a3a279a1a24234fe" target='_blank'>Improving Time Series Encoding with Noise-Aware Self-Supervised Learning and an Efficient Encoder</a></td>
+          <td>
+            Anh Duy Nguyen, Trang H. Tran, Hieu Pham, Phi-Le Nguyen, Lam M. Nguyen
+          </td>
+          <td>2023-06-11</td>
+          <td>ArXiv</td>
+          <td>3</td>
+          <td>20</td>
         </tr>
 
         <tr id="Monitoring and recognizing patterns in continuous sensing data is crucial for many practical applications. These real-world time-series data are often nonstationary, characterized by varying statistical and spectral properties over time. This poses a significant challenge in developing learning models that can effectively generalize across different distributions. In this work, based on our observation that nonstationary statistics are intrinsically linked to the phase information, we propose a time-series learning framework, PhASER. It consists of three novel elements: 1) phase augmentation that diversifies non-stationarity while preserving discriminatory semantics, 2) separate feature encoding by viewing time-varying magnitude and phase as independent modalities, and 3) feature broadcasting by incorporating phase with a novel residual connection for inherent regularization to enhance distribution invariant learning. Upon extensive evaluation on 5 datasets from human activity recognition, sleep-stage classification, and gesture recognition against 10 state-of-the-art baseline methods, we demonstrate that PhASER consistently outperforms the best baselines by an average of 5% and up to 13% in some cases. Moreover, PhASER's principles can be applied broadly to boost the generalization ability of existing time series classification models.">
@@ -569,7 +581,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2024-02-05</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>9</td>
+          <td>10</td>
         </tr>
 
         <tr id="Recent contrastive methods show significant improvement in self-supervised learning in several domains. In particular, contrastive methods are most effective where data augmentation can be easily constructed e.g. in computer vision. However, they are less successful in domains without established data transformations such as time series data. In this paper, we propose a novel self-supervised learning framework that combines contrastive learning with neural processes. It relies on recent advances in neural processes to perform time series forecasting. This allows to generate augmented versions of data by employing a set of various sampling functions and, hence, avoid manually designed augmentations. We extend conventional neural processes and propose a new contrastive loss to learn times series representations in a self-supervised setup. Therefore, unlike previous self-supervised methods, our augmentation pipeline is task-agnostic, enabling our method to perform well across various applications. In particular, a ResNet with a linear classifier trained using our approach is able to outperform state-of-the-art techniques across industrial, medical and audio datasets improving accuracy over 10% in ECG periodic data. We further demonstrate that our self-supervised representations are more efficient in the latent space, improving multiple clustering indexes and that fine-tuning our method on 10% of labels achieves results competitive to fully-supervised learning.">
@@ -584,18 +596,6 @@ <h3 data-intro='Recommendations for the article'>
           <td>13</td>
         </tr>
 
-        <tr id="Learning representations from unlabeled time series data is a challenging problem. Most existing self-supervised and unsupervised approaches in the time-series domain fall short in capturing low-and high-frequency features at the same time. As a result, the generalization ability of the learned representations remains limited. Furthermore, some of these methods employ large-scale models like transformers or rely on computationally expensive techniques such as contrastive learning. To tackle these problems, we propose a noncontrastive self-supervised learning (SSL) approach that efficiently captures low-and high-frequency features in a cost-effective manner. The proposed framework comprises a Siamese configuration of a deep neural network with two weight-sharing branches which are followed by low-and high-frequency feature extraction modules. The two branches of the proposed network allow bootstrapping of the latent representation by taking two different augmented views of raw time series data as input. The augmented views are created by applying random transformations sampled from a single set of augmentations. The low-and high-frequency feature extraction modules of the proposed network contain a combination of multilayer perceptron (MLP) and temporal convolutional network (TCN) heads, respectively, which capture the temporal dependencies from the raw input data at various scales due to the varying receptive fields. To demonstrate the robustness of our model, we performed extensive experiments and ablation studies on five real-world time-series datasets. Our method achieves state-of-art performance on all the considered datasets.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4b65c76f7f28e7ba3ea44ebb1b3e4355a148d0de" target='_blank'>Large Scale Time-Series Representation Learning via Simultaneous Low and High Frequency Feature Bootstrapping</a></td>
-          <td>
-            Vandan Gorade, Azad Singh, Deepak Mishra
-          </td>
-          <td>2022-04-24</td>
-          <td>IEEE transactions on neural networks and learning systems</td>
-          <td>5</td>
-          <td>6</td>
-        </tr>
-
   </tbody>
   <tfoot>
     <tr>
diff --git a/site/recommendations/73dd9c49f205280991826b2ea4b50344203916b4/index.html b/site/recommendations/73dd9c49f205280991826b2ea4b50344203916b4/index.html
index 46020fb6..0f547d6d 100644
--- a/site/recommendations/73dd9c49f205280991826b2ea4b50344203916b4/index.html
+++ b/site/recommendations/73dd9c49f205280991826b2ea4b50344203916b4/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:04 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:49 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-09-04</td>
           <td>Scientific Reports</td>
-          <td>2</td>
-          <td>23</td>
+          <td>3</td>
+          <td>24</td>
         </tr>
 
         <tr id="Data-driven modeling frameworks that adopt sparse regression techniques, such as sparse identification of nonlinear dynamics (SINDy) and its modifications, are developed to resolve difficulties in extracting underlying dynamics from experimental data. In contrast to neural-network-based methods, these methods are designed to obtain white-box analytical models. In this work, we incorporate the concept of SINDy and knowledge in the field of classical mechanics to identify interpretable and sparse expressions of total energy and the Lagrangian that shelters the hidden dynamics. Moreover, our method (hereafter referred as Lagrangian-SINDy) is developed to use knowledge of simple systems that form the system being analyzed to ensure the likelihood of correct results and to improve the learning pace. Lagrangian-SINDy is highly accurate in discovering interpretable dynamics via energy-related physical quantities. Its performance is validated with three popular multi-DOF nonlinear dynamical systems, namely the spherical pendulum, double pendulum and cart-pendulum system. Comparisons with other SINDy-based methods are made and Lagrangian-SINDy is found to provide the most compact analytical models.">
@@ -496,8 +496,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-01-31</td>
           <td>IEEE Robotics and Automation Letters</td>
-          <td>23</td>
-          <td>23</td>
+          <td>24</td>
+          <td>24</td>
         </tr>
 
         <tr id="In this work we analyze the effectiveness of the Sparse Identification of Nonlinear Dynamics (SINDy) technique on three benchmark datasets for nonlinear identification, to provide a better understanding of its suitability when tackling real dynamical systems. While SINDy can be an appealing strategy for pursuing physics-based learning, our analysis highlights difficulties in dealing with unobserved states and non-smooth dynamics. Due to the ubiquity of these features in real systems in general, and control applications in particular, we complement our analysis with hands-on approaches to tackle these issues in order to exploit SINDy also in these challenging contexts.">
@@ -509,7 +509,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2024-03-01</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>1</td>
+          <td>2</td>
         </tr>
 
         <tr id="In this thesis, we draw inspiration from both classical system identification and modern machine learning in order to solve estimation problems for real-world, physical systems. The main approach to estimation and learning adopted is optimization based. Concepts such as regularization will be utilized for encoding of prior knowledge and basis-function expansions will be used to add nonlinear modeling power while keeping data requirements practical. The thesis covers a wide range of applications, many inspired by applications within robotics, but also extending outside this already wide field. Usage of the proposed methods and algorithms are in many cases illustrated in the real-world applications that motivated the research. Topics covered include dynamics modeling and estimation, model-based reinforcement learning, spectral estimation, friction modeling and state estimation and calibration in robotic machining. In the work on modeling and identification of dynamics, we develop regularization strategies that allow us to incorporate prior domain knowledge into flexible, overparameterized models. We make use of classical control theory to gain insight into training and regularization while using flexible tools from modern deep learning. A particular focus of the work is to allow use of modern methods in scenarios where gathering data is associated with a high cost. In the robotics-inspired parts of the thesis, we develop methods that are practically motivated and ensure that they are implementable also outside the research setting. We demonstrate this by performing experiments in realistic settings and providing open-source implementations of all proposed methods and algorithms.">
@@ -532,7 +532,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-12-15</td>
           <td>2022 3rd International Conference on Artificial Intelligence, Robotics and Control (AIRC)</td>
-          <td>4</td>
+          <td>5</td>
           <td>5</td>
         </tr>
 
@@ -544,8 +544,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2017-11-15</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>435</td>
-          <td>65</td>
+          <td>461</td>
+          <td>67</td>
         </tr>
 
         <tr id="None">
@@ -556,7 +556,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-03-01</td>
           <td>Comput. Chem. Eng.</td>
-          <td>12</td>
+          <td>15</td>
           <td>76</td>
         </tr>
 
@@ -568,8 +568,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>202</td>
-          <td>65</td>
+          <td>221</td>
+          <td>67</td>
         </tr>
 
         <tr id="We consider the problem of learning the dynamics of a linear system when one has access to data generated by an auxiliary system that shares similar (but not identical) dynamics, in addition to data from the true system. We use a weighted least squares approach, and provide finite sample error bounds of the learned model as a function of the number of samples and various system parameters from the two systems as well as the weight assigned to the auxiliary data. We show that the auxiliary data can help to reduce the intrinsic system identification error due to noise, at the price of adding a portion of error that is due to the differences between the two system models. We further provide a data-dependent bound that is computable when some prior knowledge about the systems, such as upper bounds on noise levels and model difference, is available. This bound can also be used to determine the weight that should be assigned to the auxiliary data during the model training stage.">
@@ -581,7 +581,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2023-02-08</td>
           <td>ArXiv</td>
           <td>7</td>
-          <td>36</td>
+          <td>37</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/883547fdbd88552328a6615ec620f96e39c57018/index.html b/site/recommendations/883547fdbd88552328a6615ec620f96e39c57018/index.html
index cf36d2f1..3c6f1ceb 100644
--- a/site/recommendations/883547fdbd88552328a6615ec620f96e39c57018/index.html
+++ b/site/recommendations/883547fdbd88552328a6615ec620f96e39c57018/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:07 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:51 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-07-05</td>
           <td>Royal Society Open Science</td>
-          <td>54</td>
-          <td>31</td>
+          <td>61</td>
+          <td>32</td>
         </tr>
 
         <tr id="Sparse model identification enables nonlinear dynamical system discovery from data. However, the control of false discoveries for sparse model identification is challenging, especially in the low-data and high-noise limit. In this paper, we perform a theoretical study on ensemble sparse model discovery, which shows empirical success in terms of accuracy and robustness to noise. In particular, we analyse the bootstrapping-based sequential thresholding least-squares estimator. We show that this bootstrapping-based ensembling technique can perform a provably correct variable selection procedure with an exponential convergence rate of the error rate. In addition, we show that the ensemble sparse model discovery method can perform computationally efficient uncertainty estimation, compared to expensive Bayesian uncertainty quantification methods via MCMC. We demonstrate the convergence properties and connection to uncertainty quantification in various numerical studies on synthetic sparse linear regression and sparse model discovery. The experiments on sparse linear regression support that the bootstrapping-based sequential thresholding least-squares method has better performance for sparse variable selection compared to LASSO, thresholding least-squares, and bootstrapping-based LASSO. In the sparse model discovery experiment, we show that the bootstrapping-based sequential thresholding least-squares method can provide valid uncertainty quantification, converging to a delta measure centered around the true value with increased sample sizes. Finally, we highlight the improved robustness to hyperparameter selection under shifting noise and sparsity levels of the bootstrapping-based sequential thresholding least-squares method compared to other sparse regression methods.">
@@ -497,7 +497,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2023-01-30</td>
           <td>ArXiv</td>
           <td>11</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
 
         <tr id="None">
@@ -508,7 +508,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
 
@@ -520,8 +520,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>202</td>
-          <td>65</td>
+          <td>221</td>
+          <td>67</td>
         </tr>
 
         <tr id="We propose a fast probabilistic framework for identifying differential equations governing the dynamics of observed data. We recast the SINDy method within a Bayesian framework and use Gaussian approximations for the prior and likelihood to speed up computation. The resulting method, Bayesian-SINDy, not only quantifies uncertainty in the parameters estimated but also is more robust when learning the correct model from limited and noisy data. Using both synthetic and real-life examples such as Lynx-Hare population dynamics, we demonstrate the effectiveness of the new framework in learning correct model equations and compare its computational and data efficiency with existing methods. Because Bayesian-SINDy can quickly assimilate data and is robust against noise, it is particularly suitable for biological data and real-time system identification in control. Its probabilistic framework also enables the calculation of information entropy, laying the foundation for an active learning strategy.">
@@ -544,8 +544,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
 
         <tr id="The SINDy algorithm has been successfully used to identify the governing equations of dynamical systems from time series data. However, SINDy assumes the user has prior knowledge of the variables in the system and of a function library that can act as a basis for the system. In this paper, we demonstrate on real world data how the Augmented SINDy algorithm outperforms SINDy in the presence of system variable uncertainty. We then show SINDy can be further augmented to perform robustly when both kinds of uncertainty are present.">
@@ -560,6 +560,18 @@ <h3 data-intro='Recommendations for the article'>
           <td>0</td>
         </tr>
 
+        <tr id="Active learning algorithms strategically select informative input samples, aiming to enhance understanding and optimize sample efficiency for dynamical systems. The efficacy of machine learning (ML) models with uncertainty quantification in identifying dynamical systems with different characteristics has not yet been explored. This paper investigates four ML models: Gaussian processes (GPs) with radial basis function (RBF), GPs with Matérn kernels, Bayesian neural networks (NNs) with Monte Carlo dropout(MCD), and ensemble neural networks (ENNs). Both continuous and non-continuous learning strategies are employed for all models. Five well known dynamical systems with different characteristics, state dimensionalities, and dynamical complexities are simulated to test the performance of the models along the metrics of test RMSE as well as samples and run time until convergence. Through systematic experimentation, this paper provides guidance for researchers selecting models and algorithms in active learning for system identification, with insights into their strengths and limitations. Gaussian Processes with RBF kernels are shown to be the best overall choice while ENNs with Bayesian inference emerge as a scalable alternative, showing promise for high-dimensional tasks. Employing continuous training has proven to enhance the performance of Gaussian processes with RBF kernels, as well as ENNs, resulting in significant improvements in both sample and run-time efficiency.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/4a3887d48d70464cf825d7d6d8834e36e77b5d04" target='_blank'>Comparative Analysis of Uncertainty Quantification Models in Active Learning for Efficient System Identification of Dynamical Systems</a></td>
+          <td>
+            Hans Mertens, Frances Zhu
+          </td>
+          <td>2024-08-28</td>
+          <td>2024 IEEE 20th International Conference on Automation Science and Engineering (CASE)</td>
+          <td>0</td>
+          <td>0</td>
+        </tr>
+
         <tr id="We consider the data-driven discovery of governing equations from time-series data in the limit of high noise. The algorithms developed describe an extensive toolkit of methods for circumventing the deleterious effects of noise in the context of the <italic>sparse identification of nonlinear dynamics</italic> (SINDy) framework. We offer two primary contributions, both focused on noisy data acquired from a system <inline-formula> <tex-math notation="LaTeX">$\dot { \boldsymbol x} = { \boldsymbol f} ({ \boldsymbol x})$ </tex-math></inline-formula>. First, we propose, for use in high-noise settings, an extensive toolkit of critically enabling extensions for the SINDy regression method, to progressively cull functionals from an over-complete library and yield a set of sparse equations that regress to the derivate <inline-formula> <tex-math notation="LaTeX">$\dot { \boldsymbol {x}}$ </tex-math></inline-formula>. This toolkit includes: (regression step) weight timepoints based on estimated noise, use ensembles to estimate coefficients, and regress using FFTs; (culling step) leverage linear dependence of functionals, and restore and protect culled functionals based on Figures of Merit (FoMs). In a novel Assessment step, we define FoMs that compare model predictions to the original time-series (i.e., <inline-formula> <tex-math notation="LaTeX">${ \boldsymbol x}(t)$ </tex-math></inline-formula> rather than <inline-formula> <tex-math notation="LaTeX">$\dot { \boldsymbol {x}}(t)$ </tex-math></inline-formula>). These innovations can extract sparse governing equations and coefficients from high-noise time-series data (e.g., 300% added noise). For example, it discovers the correct sparse libraries in the Lorenz system, with median coefficient estimate errors equal to 1%−3% (for 50% noise), 6%−8% (for 100% noise), and 23%−25% (for 300% noise). The enabling modules in the toolkit are combined into a single method, but the individual modules can be tactically applied in other equation discovery methods (SINDy or not) to improve results on high-noise data. Second, we propose a technique, applicable to any model discovery method based on <inline-formula> <tex-math notation="LaTeX">$\dot { \boldsymbol x} = { \boldsymbol f} ({ \boldsymbol x})$ </tex-math></inline-formula>, to assess the accuracy of a discovered model in the context of non-unique solutions due to noisy data. Currently, this non-uniqueness can obscure a discovered model’s accuracy and thus a discovery method’s effectiveness. We describe a technique that uses linear dependencies among functionals to transform a discovered model into an equivalent form that is closest to the <italic>true</italic> model, enabling more accurate assessment of a discovered model’s correctness.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/a9803137946f0dc8997f23de925d5aec6e92dfb9" target='_blank'>A Toolkit for Data-Driven Discovery of Governing Equations in High-Noise Regimes</a></td>
@@ -569,7 +581,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2021-11-08</td>
           <td>IEEE Access</td>
           <td>16</td>
-          <td>31</td>
+          <td>32</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271/index.html b/site/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271/index.html
index 31cdbbb9..710caf79 100644
--- a/site/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271/index.html
+++ b/site/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:57 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:43 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -476,18 +476,6 @@ <h3 data-intro='Recommendations for the article'>
   </thead>
   <tbody>
 
-        <tr id="Many dynamical systems of interest are nonlinear, with examples in turbulence, epidemiology, neuroscience, and finance, making them difficult to control using linear approaches. Model predictive control (MPC) is a powerful model-based optimization technique that enables the control of such nonlinear systems with constraints. However, modern systems often lack computationally tractable models, motivating the use of system identification techniques to learn accurate and efficient models for real-time control. In this tutorial article, we review emerging data-driven methods for model discovery and how they are used for nonlinear MPC. In particular, we focus on the sparse identification of nonlinear dynamics (SINDy) algorithm and show how it may be used with MPC on an infectious disease control example. We compare the performance against MPC based on a linear dynamic mode decomposition (DMD) model. Code is provided to run the tutorial examples and may be modified to extend this data-driven control framework to arbitrary nonlinear systems.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/700552dde778bd1f6b76f7237e9b561597b859ba" target='_blank'>SINDy with Control: A Tutorial</a></td>
-          <td>
-            Urban Fasel, E. Kaiser, J. Kutz, Bingni W. Brunton, S. Brunton
-          </td>
-          <td>2021-08-30</td>
-          <td>2021 60th IEEE Conference on Decision and Control (CDC)</td>
-          <td>48</td>
-          <td>65</td>
-        </tr>
-
         <tr id="In this work we analyze the effectiveness of the Sparse Identification of Nonlinear Dynamics (SINDy) technique on three benchmark datasets for nonlinear identification, to provide a better understanding of its suitability when tackling real dynamical systems. While SINDy can be an appealing strategy for pursuing physics-based learning, our analysis highlights difficulties in dealing with unobserved states and non-smooth dynamics. Due to the ubiquity of these features in real systems in general, and control applications in particular, we complement our analysis with hands-on approaches to tackle these issues in order to exploit SINDy also in these challenging contexts.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/99877062e70fae24c531d58aa880fa443ec10c9b" target='_blank'>SINDy vs Hard Nonlinearities and Hidden Dynamics: a Benchmarking Study</a></td>
@@ -497,7 +485,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2024-03-01</td>
           <td>ArXiv</td>
           <td>1</td>
-          <td>1</td>
+          <td>2</td>
         </tr>
 
         <tr id="Data-driven modeling frameworks that adopt sparse regression techniques, such as sparse identification of nonlinear dynamics (SINDy) and its modifications, are developed to resolve difficulties in extracting underlying dynamics from experimental data. In contrast to neural-network-based methods, these methods are designed to obtain white-box analytical models. In this work, we incorporate the concept of SINDy and knowledge in the field of classical mechanics to identify interpretable and sparse expressions of total energy and the Lagrangian that shelters the hidden dynamics. Moreover, our method (hereafter referred as Lagrangian-SINDy) is developed to use knowledge of simple systems that form the system being analyzed to ensure the likelihood of correct results and to improve the learning pace. Lagrangian-SINDy is highly accurate in discovering interpretable dynamics via energy-related physical quantities. Its performance is validated with three popular multi-DOF nonlinear dynamical systems, namely the spherical pendulum, double pendulum and cart-pendulum system. Comparisons with other SINDy-based methods are made and Lagrangian-SINDy is found to provide the most compact analytical models.">
@@ -508,8 +496,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-01-31</td>
           <td>IEEE Robotics and Automation Letters</td>
-          <td>23</td>
-          <td>23</td>
+          <td>24</td>
+          <td>24</td>
         </tr>
 
         <tr id="Accurately modelling the nonlinear dynamics of a system from measurement data is a challenging yet vital topic. The sparse identification of nonlinear dynamics (SINDy) algorithm is one approach to discover dynamical systems models from data. Although extensions have been developed to identify implicit dynamics, or dynamics described by rational functions, these extensions are extremely sensitive to noise. In this work, we develop SINDy-PI (parallel, implicit), a robust variant of the SINDy algorithm to identify implicit dynamics and rational nonlinearities. The SINDy-PI framework includes multiple optimization algorithms and a principled approach to model selection. We demonstrate the ability of this algorithm to learn implicit ordinary and partial differential equations and conservation laws from limited and noisy data. In particular, we show that the proposed approach is several orders of magnitude more noise robust than previous approaches, and may be used to identify a class of ODE and PDE dynamics that were previously unattainable with SINDy, including for the double pendulum dynamics and simplified model for the Belousov–Zhabotinsky (BZ) reaction.">
@@ -520,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>202</td>
-          <td>65</td>
+          <td>221</td>
+          <td>67</td>
         </tr>
 
         <tr id="Modern societies have an abundance of data yet good system models are rare. Unfortunately, many of the current system identification and machine learning techniques fail to generalize outside of the training set, producing models that violate basic physical laws. This work proposes a novel method for the Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI). SINDy-SI is an iterative method that uses Sum-of-Squares (SOS) programming to learn optimally fitted models while guaranteeing that the learned model satisfies side information, such as symmetries and physical laws. Guided by the principle of Occam's razor, that the simplest or most regularized best fitted model is typically the superior choice, during each iteration SINDy-SI prunes the basis functions associated with small coefficients, yielding a sparse dynamical model upon termination. Through several numerical experiments we will show how the combination of side information constraints and sparse polynomial representation cultivates dynamical models that obey known physical laws while displaying impressive generalized performance beyond the training set.">
@@ -536,64 +524,64 @@ <h3 data-intro='Recommendations for the article'>
           <td>1</td>
         </tr>
 
-        <tr id="Automated data-driven modeling, the process of directly discovering the governing equations of a system from data, is increasingly being used across the scientific community. PySINDy is a Python package that provides tools for applying the sparse identification of nonlinear dynamics (SINDy) approach to data-driven model discovery. In this major update to PySINDy, we implement several advanced features that enable the discovery of more general differential equations from noisy and limited data. The library of candidate terms is extended for the identification of actuated systems, partial differential equations (PDEs), and implicit differential equations. Robust formulations, including the integral form of SINDy and ensembling techniques, are also implemented to improve performance for real-world data. Finally, we provide a range of new optimization algorithms, including several sparse regression techniques and algorithms to enforce and promote inequality constraints and stability. Together, these updates enable entirely new SINDy model discovery capabilities that have not been reported in the literature, such as constrained PDE identification and ensembling with different sparse regression optimizers.">
+        <tr id="The sparse identification of nonlinear dynamics (SINDy) has been established as an effective technique to produce interpretable models of dynamical systems from time-resolved state data via sparse regression. However, to model parameterized systems, SINDy requires data from transient trajectories for various parameter values over the range of interest, which are typically difficult to acquire experimentally. In this work, we extend SINDy to be able to leverage data on fixed points and/or limit cycles to reduce the number of transient trajectories needed for successful system identification. To achieve this, we incorporate the data on these attractors at various parameter values as constraints in the optimization problem. First, we show that enforcing these as hard constraints leads to an ill-conditioned regression problem due to the large number of constraints. Instead, we implement soft constraints by modifying the cost function to be minimized. This leads to the formulation of a multi-objective sparse regression problem where we simultaneously seek to minimize the error of the fit to the transients trajectories and to the data on attractors, while penalizing the number of terms in the model. Our extension, demonstrated on several numerical examples, is more robust to noisy measurements and requires substantially less training data than the original SINDy method to correctly identify a parameterized dynamical system.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/433ba35ced5d3f47fa2688c81a60363a74193449" target='_blank'>PySINDy: A comprehensive Python package for robust sparse system identification</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/0c03c126b4d641a81099470f03a7d5215a2a6820" target='_blank'>Multi-objective SINDy for parameterized model discovery from single transient trajectory data</a></td>
           <td>
-            A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton
+            Javier A. Lemus, Benjamin Herrmann
           </td>
-          <td>2021-11-12</td>
-          <td>J. Open Source Softw.</td>
-          <td>111</td>
-          <td>65</td>
+          <td>2024-05-14</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>0</td>
         </tr>
 
-        <tr id="None">
+        <tr id="Significance Understanding dynamic constraints and balances in nature has facilitated rapid development of knowledge and enabled technology, including aircraft, combustion engines, satellites, and electrical power. This work develops a novel framework to discover governing equations underlying a dynamical system simply from data measurements, leveraging advances in sparsity techniques and machine learning. The resulting models are parsimonious, balancing model complexity with descriptive ability while avoiding overfitting. There are many critical data-driven problems, such as understanding cognition from neural recordings, inferring climate patterns, determining stability of financial markets, predicting and suppressing the spread of disease, and controlling turbulence for greener transportation and energy. With abundant data and elusive laws, data-driven discovery of dynamics will continue to play an important role in these efforts. Extracting governing equations from data is a central challenge in many diverse areas of science and engineering. Data are abundant whereas models often remain elusive, as in climate science, neuroscience, ecology, finance, and epidemiology, to name only a few examples. In this work, we combine sparsity-promoting techniques and machine learning with nonlinear dynamical systems to discover governing equations from noisy measurement data. The only assumption about the structure of the model is that there are only a few important terms that govern the dynamics, so that the equations are sparse in the space of possible functions; this assumption holds for many physical systems in an appropriate basis. In particular, we use sparse regression to determine the fewest terms in the dynamic governing equations required to accurately represent the data. This results in parsimonious models that balance accuracy with model complexity to avoid overfitting. We demonstrate the algorithm on a wide range of problems, from simple canonical systems, including linear and nonlinear oscillators and the chaotic Lorenz system, to the fluid vortex shedding behind an obstacle. The fluid example illustrates the ability of this method to discover the underlying dynamics of a system that took experts in the community nearly 30 years to resolve. We also show that this method generalizes to parameterized systems and systems that are time-varying or have external forcing.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/899995e30820d657b2e2feb791025cd6b2027e22" target='_blank'>Discovering sparse interpretable dynamics from partial observations</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/5d150cec2775f9bc863760448f14104cc8f42368" target='_blank'>Discovering governing equations from data by sparse identification of nonlinear dynamical systems</a></td>
           <td>
-            Peter Y. Lu, Joan Ariño Bernad, M. Soljačić
+            S. Brunton, J. Proctor, J. Kutz
           </td>
-          <td>2021-07-22</td>
-          <td>Communications Physics</td>
-          <td>17</td>
-          <td>94</td>
+          <td>2015-09-11</td>
+          <td>Proceedings of the National Academy of Sciences</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
 
-        <tr id="Authors of papers retaincopyright and release the workunder a Creative CommonsAttribution 4.0 InternationalLicense (CC-BY)">
+        <tr id="Automated data-driven modeling, the process of directly discovering the governing equations of a system from data, is increasingly being used across the scientific community. PySINDy is a Python package that provides tools for applying the sparse identification of nonlinear dynamics (SINDy) approach to data-driven model discovery. In this major update to PySINDy, we implement several advanced features that enable the discovery of more general differential equations from noisy and limited data. The library of candidate terms is extended for the identification of actuated systems, partial differential equations (PDEs), and implicit differential equations. Robust formulations, including the integral form of SINDy and ensembling techniques, are also implemented to improve performance for real-world data. Finally, we provide a range of new optimization algorithms, including several sparse regression techniques and algorithms to enforce and promote inequality constraints and stability. Together, these updates enable entirely new SINDy model discovery capabilities that have not been reported in the literature, such as constrained PDE identification and ensembling with different sparse regression optimizers.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/85d5e9aea7235bf6aa67d5febc4013923a77d4de" target='_blank'>PySINDy: A Python package for the sparse identification of nonlinear dynamical systems from data</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/433ba35ced5d3f47fa2688c81a60363a74193449" target='_blank'>PySINDy: A comprehensive Python package for robust sparse system identification</a></td>
           <td>
-            Brian M. de Silva, Kathleen P. Champion, M. Quade, Jean-Christophe Loiseau, J. Kutz, S. Brunton
+            A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton
           </td>
-          <td>2020-05-18</td>
+          <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>129</td>
-          <td>65</td>
+          <td>119</td>
+          <td>67</td>
         </tr>
 
-        <tr id="The vast majority of systems of practical interest are characterised by nonlinear dynamics. This renders the control and optimization of such systems a complex task due to their nonlinear behaviour. Additionally, standard methods such as linearizing around a fixed point may not be an effective strategy for many systems, thus requiring an alternative approach. For this reason, we propose a new deep learning framework to discover a transformation of a nonlinear dynamical system to an equivalent higher dimensional linear representation. We demonstrate that the resulting learned linear representation accurately captures the dynamics of the original system for a wider range of conditions than standard linearization. As a result of this, we show that the learned linear model can subsequently be used for the successful control of the original system. We demonstrate this by applying the proposed framework to two examples; one from the literature and a more complex example in the form of a Continuous Stirred Tank Reactor (CSTR).">
+        <tr id="Identifying dynamical systems characterized by nonlinear parameters presents significant challenges in deriving mathematical models that enhance understanding of physics. Traditional methods, such as Sparse Identification of Nonlinear Dynamics (SINDy) and symbolic regression, can extract governing equations from observational data; however, they also come with distinct advantages and disadvantages. This paper introduces a novel method within the SINDy framework, termed ADAM-SINDy, which synthesizes the strengths of established approaches by employing the ADAM optimization algorithm. This facilitates the simultaneous optimization of nonlinear parameters and coefficients associated with nonlinear candidate functions, enabling precise parameter estimation without requiring prior knowledge of nonlinear characteristics such as trigonometric frequencies, exponential bandwidths, or polynomial exponents, thereby addressing a key limitation of SINDy. Through an integrated global optimization, ADAM-SINDy dynamically adjusts all unknown variables in response to data, resulting in an adaptive identification procedure that reduces the sensitivity to the library of candidate functions. The performance of the ADAM-SINDy methodology is demonstrated across a spectrum of dynamical systems, including benchmark coupled nonlinear ordinary differential equations such as oscillators, chaotic fluid flows, reaction kinetics, pharmacokinetics, as well as nonlinear partial differential equations (wildfire transport). The results demonstrate significant improvements in identifying parameterized dynamical systems and underscore the importance of concurrently optimizing all parameters, particularly those characterized by nonlinear parameters. These findings highlight the potential of ADAM-SINDy to extend the applicability of the SINDy framework in addressing more complex challenges in dynamical system identification.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c309d30d468a48f5c5d40e402af5881451bb6117" target='_blank'>Learning Linear Representations of Nonlinear Dynamics Using Deep Learning</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/0c327ed253a2c7b658940860bcf0a9e077d634b6" target='_blank'>ADAM-SINDy: An Efficient Optimization Framework for Parameterized Nonlinear Dynamical System Identification</a></td>
           <td>
-            Akhil Ahmed, E. A. Rio-Chanona, Mehmet Mercangöz
+            Siva Viknesh, Younes Tatari, Amirhossein Arzani
           </td>
-          <td>2022-04-03</td>
+          <td>2024-10-21</td>
           <td>ArXiv</td>
-          <td>3</td>
-          <td>19</td>
+          <td>0</td>
+          <td>24</td>
         </tr>
 
-        <tr id="None">
+        <tr id="We develop a principled mathematical framework for controlling nonlinear, networked dynamical systems. Our method integrates dimensionality reduction, bifurcation theory, and emerging model discovery tools to find low-dimensional subspaces where feed-forward control can be used to manipulate a system to a desired outcome. The method leverages the fact that many high-dimensional networked systems have many fixed points, allowing for the computation of control signals that will move the system between any pair of fixed points. The sparse identification of nonlinear dynamics (SINDy) algorithm is used to fit a nonlinear dynamical system to the evolution on the dominant, low-rank subspace. This then allows us to use bifurcation theory to find collections of constant control signals that will produce the desired objective path for a prescribed outcome. Specifically, we can destabilize a given fixed point while making the target fixed point an attractor. The discovered control signals can be easily projected back to the original high-dimensional state and control space. We illustrate our nonlinear control procedure on established bistable, low-dimensional biological systems, showing how control signals are found that generate switches between the fixed points. We then demonstrate our control procedure for high-dimensional systems on random high-dimensional networks and Hopfield memory networks.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/acd85b948087067938680356a0cb7597de12a25d" target='_blank'>Data-based modeling and control of nonlinear process systems using sparse identification: An overview of recent results</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/4b849d45b437a09d1ea5321a304c5bc411bf3de4" target='_blank'>Nonlinear Control of Networked Dynamical Systems</a></td>
           <td>
-            Fahim Abdullah, P. Christofides
+            Megan Morrison, Nathan Kutz
           </td>
-          <td>2023-03-01</td>
-          <td>Comput. Chem. Eng.</td>
-          <td>12</td>
-          <td>76</td>
+          <td>2020-06-09</td>
+          <td>IEEE Transactions on Network Science and Engineering</td>
+          <td>10</td>
+          <td>5</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da/index.html b/site/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da/index.html
index 3f75f7bb..45704074 100644
--- a/site/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da/index.html
+++ b/site/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:08 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:54 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -485,7 +485,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2018-07-14</td>
           <td>ArXiv</td>
           <td>9</td>
-          <td>65</td>
+          <td>67</td>
         </tr>
 
         <tr id="Sparse regression models are increasingly prevalent due to their ease of interpretability and superior out-of-sample performance. However, the exact model of sparse regression with an $\ell_0$ constraint restricting the support of the estimators is a challenging (\NP-hard) non-convex optimization problem. In this paper, we derive new strong convex relaxations for sparse regression. These relaxations are based on the ideal (convex-hull) formulations for rank-one quadratic terms with indicator variables. The new relaxations can be formulated as semidefinite optimization problems in an extended space and are stronger and more general than the state-of-the-art formulations, including the perspective reformulation and formulations with the reverse Huber penalty and the minimax concave penalty functions. Furthermore, the proposed rank-one strengthening can be interpreted as a \textit{non-separable, non-convex, unbiased} sparsity-inducing regularizer, which dynamically adjusts its penalty according to the shape of the error function without inducing bias for the sparse solutions. In our computational experiments with benchmark datasets, the proposed conic formulations are solved within seconds and result in near-optimal solutions (with 0.4\% optimality gap) for non-convex $\ell_0$-problems. Moreover, the resulting estimators also outperform alternative convex approaches from a statistical perspective, achieving high prediction accuracy and good interpretability.">
@@ -496,8 +496,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2019-01-29</td>
           <td>ArXiv</td>
-          <td>49</td>
-          <td>35</td>
+          <td>48</td>
+          <td>36</td>
         </tr>
 
         <tr id="In the context of sparse recovery, it is known that most of existing regularizers such as $\ell_1$ suffer from some bias incurred by some leading entries (in magnitude) of the associated vector. To neutralize this bias, we propose a class of models with partial regularizers for recovering a sparse solution of a linear system. We show that every local minimizer of these models is sufficiently sparse or the magnitude of all its nonzero entries is above a uniform constant depending only on the data of the linear system. Moreover, for a class of partial regularizers, any global minimizer of these models is a sparsest solution to the linear system. We also establish some sufficient conditions for local or global recovery of the sparsest solution to the linear system, among which one of the conditions is weaker than the best known restricted isometry property (RIP) condition for sparse recovery by $\ell_1$. In addition, a first-order feasible augmented Lagrangian (FAL) method is proposed for solving these models, in which each subproblem is solved by a nonmonotone proximal gradient (NPG) method. Despite the complication of the partial regularizers, we show that each proximal subproblem in NPG can be solved as a certain number of one-dimensional optimization problems, which usually have a closed-form solution. We also show that any accumulation point of the sequence generated by FAL is a first-order stationary point of the models. Numerical results on compressed sensing and sparse logistic regression demonstrate that the proposed models substantially outperform the widely used ones in the literature in terms of solution quality.">
@@ -508,7 +508,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2015-11-23</td>
           <td>ArXiv</td>
-          <td>36</td>
+          <td>37</td>
           <td>32</td>
         </tr>
 
@@ -547,7 +547,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2021-10-24</td>
           <td>ArXiv</td>
           <td>2</td>
-          <td>16</td>
+          <td>17</td>
         </tr>
 
         <tr id="This paper addresses the robust reconstruction problem of a sparse signal from compressed measurements. We propose a robust formulation for sparse reconstruction that employs the <inline-formula><tex-math notation="LaTeX">$\ell _1$ </tex-math></inline-formula>-norm as the loss function for the residual error and utilizes a generalized nonconvex penalty for sparsity inducing. The <inline-formula><tex-math notation="LaTeX">$\ell _1$</tex-math></inline-formula> -loss is less sensitive to outliers in the measurements than the popular <inline-formula><tex-math notation="LaTeX"> $\ell _2$</tex-math></inline-formula>-loss, while the nonconvex penalty has the capability of ameliorating the bias problem of the popular convex LASSO penalty and thus can yield more accurate recovery. To solve this nonconvex and nonsmooth minimization formulation efficiently, we propose a first-order algorithm based on alternating direction method of multipliers. A smoothing strategy on the <inline-formula><tex-math notation="LaTeX">$\ell _1$</tex-math> </inline-formula>-loss function has been used in deriving the new algorithm to make it convergent. Further, a sufficient condition for the convergence of the new algorithm has been provided for generalized nonconvex regularization. In comparison with several state-of-the-art algorithms, the new algorithm showed better performance in numerical experiments in recovering sparse signals and compressible images. The new algorithm scales well for large-scale problems, as often encountered in image processing.">
@@ -558,8 +558,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2017-03-23</td>
           <td>IEEE Transactions on Computational Imaging</td>
-          <td>85</td>
-          <td>30</td>
+          <td>87</td>
+          <td>31</td>
         </tr>
 
         <tr id="None">
diff --git a/site/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524/index.html b/site/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524/index.html
index 707ed9a0..e9ea192e 100644
--- a/site/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524/index.html
+++ b/site/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:33 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:25 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,7 +484,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-04-23</td>
           <td>ArXiv</td>
-          <td>5</td>
+          <td>7</td>
           <td>50</td>
         </tr>
 
@@ -495,8 +495,8 @@ <h3 data-intro='Recommendations for the article'>
             Ming Jin, Huan Yee Koh, Qingsong Wen, Daniele Zambon, C. Alippi, G. I. Webb, Irwin King, Shirui Pan
           </td>
           <td>2023-07-07</td>
-          <td>IEEE transactions on pattern analysis and machine intelligence</td>
-          <td>64</td>
+          <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
+          <td>82</td>
           <td>50</td>
         </tr>
 
@@ -509,7 +509,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2023-12-27</td>
           <td>2024 IEEE 40th International Conference on Data Engineering (ICDE)</td>
           <td>0</td>
-          <td>12</td>
+          <td>13</td>
         </tr>
 
         <tr id="None">
@@ -520,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-09-25</td>
           <td>International Journal of Data Science and Analytics</td>
-          <td>0</td>
+          <td>2</td>
           <td>6</td>
         </tr>
 
@@ -532,7 +532,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-02-22</td>
           <td>ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)</td>
-          <td>5</td>
+          <td>7</td>
           <td>42</td>
         </tr>
 
@@ -544,8 +544,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-06-30</td>
           <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
-          <td>7</td>
-          <td>21</td>
+          <td>8</td>
+          <td>22</td>
         </tr>
 
         <tr id="Multivariate time series forecasting has long received significant attention in real-world applications, such as energy consumption and traffic prediction. While recent methods demonstrate good forecasting abilities, they have three fundamental limitations. (i). <italic>Discrete neural architectures:</italic> Interlacing individually parameterized spatial and temporal blocks to encode rich underlying patterns leads to discontinuous latent state trajectories and higher forecasting numerical errors. (ii). <italic>High complexity:</italic> Discrete approaches complicate models with dedicated designs and redundant parameters, leading to higher computational and memory overheads. (iii). <italic>Reliance on graph priors:</italic> Relying on predefined static graph structures limits their effectiveness and practicability in real-world applications. In this paper, we address all the above limitations by proposing a continuous model to forecast <underline><bold>M</bold></underline>ultivariate <underline><bold>T</bold></underline>ime series with dynamic <underline><bold>G</bold></underline>raph neural <underline><bold>O</bold></underline>rdinary <underline><bold>D</bold></underline>ifferential <underline><bold>E</bold></underline>quations (<monospace>MTGODE</monospace>). Specifically, we first abstract multivariate time series into dynamic graphs with time-evolving node features and unknown graph structures. Then, we design and solve a neural ODE to complement missing graph topologies and unify both spatial and temporal message passing, allowing deeper graph propagation and fine-grained temporal information aggregation to characterize stable and precise latent spatial-temporal dynamics. Our experiments demonstrate the superiorities of <monospace>MTGODE</monospace> from various perspectives on five time series benchmark datasets.">
@@ -556,8 +556,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-02-17</td>
           <td>IEEE Transactions on Knowledge and Data Engineering</td>
-          <td>60</td>
-          <td>43</td>
+          <td>69</td>
+          <td>44</td>
         </tr>
 
         <tr id="Existing multivariate time series anomaly detection methods utilize graph neural networks to model inter-sensor dependencies and achieve satisfactory results. However, these methods often employ a predefined or self-learning static adjacency matrix for graph information aggregation, disregarding the dynamic characteristics of inter-sensor dependencies in the real world, which change over time and are observable at different time scales as dynamic changing graph structure information. Furthermore, existing techniques struggle to capture long-term temporal contextual relationships. To address these issues, we propose a Dynamic Graph Learning with Long and Short-term time model (DGL-LS) for multivariate time series anomaly detection. The network integrates a dynamic graph structure learning method among sensors in multi-time mode and long- and short-term temporal context information to enhance the prediction of multivariate time series anomalies. We propose a strategy for learning dynamic graphs that aggregates sensor node representations at various time intervals, and then uses an attention-based gating unit to evolve the node dynamic characterization and construct a series of adjacency matrices that describe the dynamic changes in graph structure over time. Moreover, we formulate a long-short time convolution to capture the long-term and short-term temporal dependencies within the time series. We demonstrate the superiority of our approach over other state-of-the-art methods by conducting extensive experiments on three publicly available datasets.">
diff --git a/site/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a/index.html b/site/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a/index.html
index 77c41e71..3ceb2a75 100644
--- a/site/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a/index.html
+++ b/site/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:16 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:11 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>54</td>
+          <td>1</td>
+          <td>55</td>
         </tr>
 
         <tr id="Accurate forecasting of multivariate time series is an extensively studied subject in finance, transportation, and computer science. Fully mining the correlation and causation between the variables in a multivariate time series exhibits noticeable results in improving the performance of a time series model. Recently, some models have explored the dependencies between variables through end-to-end graph structure learning without the need for predefined graphs. However, current models do not incorporate the trade-off between efficiency and flexibility and lack the guidance of domain knowledge in the design of graph structure learning algorithms. This paper alleviates the above issues by proposing Balanced Graph Structure Learning for Forecasting (BGSLF), a novel deep learning model that joins graph structure learning and forecasting. Technically, BGSLF leverages the spatial information into convolutional operations and extracts temporal dynamics using the diffusion convolutional recurrent network. The proposed framework balance the trade-off between efficiency and flexibility by introducing Multi-Graph Generation Network (MGN) and Graph Selection Module. In addition, a method named Smooth Sparse Unit (SSU) is designed to sparse the learned graph structures, which conforms to the sparse spatial correlations in the real world. Extensive experiments on four real-world datasets demonstrate that our model achieves state-of-the-art performances with minor trainable parameters. Code will be made publicly available.">
@@ -497,7 +497,7 @@ <h3 data-intro='Recommendations for the article'>
           <td>2022-01-24</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>10</td>
+          <td>11</td>
         </tr>
 
         <tr id="Outstanding achievements of graph neural networks for spatiotemporal time series analysis show that relational constraints introduce an effective inductive bias into neural forecasting architectures. Often, however, the relational information characterizing the underlying data-generating process is unavailable and the practitioner is left with the problem of inferring from data which relational graph to use in the subsequent processing stages. We propose novel, principled - yet practical - probabilistic score-based methods that learn the relational dependencies as distributions over graphs while maximizing end-to-end the performance at task. The proposed graph learning framework is based on consolidated variance reduction techniques for Monte Carlo score-based gradient estimation, is theoretically grounded, and, as we show, effective in practice. In this paper, we focus on the time series forecasting problem and show that, by tailoring the gradient estimators to the graph learning problem, we are able to achieve state-of-the-art performance while controlling the sparsity of the learned graph and the computational scalability. We empirically assess the effectiveness of the proposed method on synthetic and real-world benchmarks, showing that the proposed solution can be used as a stand-alone graph identification procedure as well as a graph learning component of an end-to-end forecasting architecture.">
@@ -508,7 +508,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-05-26</td>
           <td>J. Mach. Learn. Res.</td>
-          <td>11</td>
+          <td>14</td>
           <td>50</td>
         </tr>
 
@@ -532,8 +532,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-05-24</td>
           <td>Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining</td>
-          <td>981</td>
-          <td>55</td>
+          <td>1082</td>
+          <td>57</td>
         </tr>
 
         <tr id="Graph Neural Networks (GNN) have gained significant traction in the forecasting domain, especially for their capacity to simultaneously account for intra-series temporal correlations and inter-series relationships. This paper introduces a novel Hierarchical GNN (DeepHGNN) framework, explicitly designed for forecasting in complex hierarchical structures. The uniqueness of DeepHGNN lies in its innovative graph-based hierarchical interpolation and an end-to-end reconciliation mechanism. This approach ensures forecast accuracy and coherence across various hierarchical levels while sharing signals across them, addressing a key challenge in hierarchical forecasting. A critical insight in hierarchical time series is the variance in forecastability across levels, with upper levels typically presenting more predictable components. DeepHGNN capitalizes on this insight by pooling and leveraging knowledge from all hierarchy levels, thereby enhancing the overall forecast accuracy. Our comprehensive evaluation set against several state-of-the-art models confirm the superior performance of DeepHGNN. This research not only demonstrates DeepHGNN's effectiveness in achieving significantly improved forecast accuracy but also contributes to the understanding of graph-based methods in hierarchical time series forecasting.">
@@ -556,7 +556,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-11-10</td>
           <td>ArXiv</td>
-          <td>32</td>
+          <td>49</td>
           <td>6</td>
         </tr>
 
diff --git a/site/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9/index.html b/site/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9/index.html
index ad8f4a32..9efdc9fb 100644
--- a/site/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9/index.html
+++ b/site/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:08 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:07 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,10 +484,22 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-12-23</td>
           <td>ACM Transactions on Knowledge Discovery from Data</td>
-          <td>1</td>
+          <td>3</td>
           <td>15</td>
         </tr>
 
+        <tr id="In recent years, spatio-temporal graph neural networks (GNNs) have attracted considerable interest in the field of time series analysis, due to their ability to capture dependencies among variables and across time points. The objective of the presented systematic literature review is hence to provide a comprehensive overview of the various modeling approaches and application domains of GNNs for time series classification and forecasting. A database search was conducted, and over 150 journal papers were selected for a detailed examination of the current state-of-the-art in the field. This examination is intended to offer to the reader a comprehensive collection of proposed models, links to related source code, available datasets, benchmark models, and fitting results. All this information is hoped to assist researchers in future studies. To the best of our knowledge, this is the first systematic literature review presenting a detailed comparison of the results of current spatio-temporal GNN models in different domains. In addition, in its final part this review discusses current limitations and challenges in the application of spatio-temporal GNNs, such as comparability, reproducibility, explainability, poor information capacity, and scalability.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/e6544d542d13247f620834689ca2378cb7fb04f8" target='_blank'>A Systematic Literature Review of Spatio-Temporal Graph Neural Network Models for Time Series Forecasting and Classification</a></td>
+          <td>
+            Flavio Corradini, Marco Gori, Carlo Lucheroni, Marco Piangerelli, Martina Zannotti
+          </td>
+          <td>2024-10-29</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>12</td>
+        </tr>
+
         <tr id="Time series forecasting lies at the core of important real-world applications in many fields of science and engineering. The abundance of large time series datasets that consist of complex patterns and long-term dependencies has led to the development of various neural network architectures. Graph neural network approaches, which jointly learn a graph structure based on the correlation of raw values of multivariate time series while forecasting, have recently seen great success. However, such solutions are often costly to train and difficult to scale. In this paper, we propose TimeGNN, a method that learns dynamic temporal graph representations that can capture the evolution of inter-series patterns along with the correlations of multiple series. TimeGNN achieves inference times 4 to 80 times faster than other state-of-the-art graph-based methods while achieving comparable forecasting performance">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/d9916caa8da69496ecc8e0b1c229c73a60d6f861" target='_blank'>TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting</a></td>
@@ -496,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>54</td>
+          <td>1</td>
+          <td>55</td>
         </tr>
 
         <tr id="Graph-based deep learning methods have become popular tools to process collections of correlated time series. Differently from traditional multivariate forecasting methods, neural graph-based predictors take advantage of pairwise relationships by conditioning forecasts on a (possibly dynamic) graph spanning the time series collection. The conditioning can take the form of an architectural inductive bias on the neural forecasting architecture, resulting in a family of deep learning models called spatiotemporal graph neural networks. Such relational inductive biases enable the training of global forecasting models on large time-series collections, while at the same time localizing predictions w.r.t. each element in the set (i.e., graph nodes) by accounting for local correlations among them (i.e., graph edges). Indeed, recent theoretical and practical advances in graph neural networks and deep learning for time series forecasting make the adoption of such processing frameworks appealing and timely. However, most of the studies in the literature focus on proposing variations of existing neural architectures by taking advantage of modern deep learning practices, while foundational and methodological aspects have not been subject to systematic investigation. To fill the gap, this paper aims to introduce a comprehensive methodological framework that formalizes the forecasting problem and provides design principles for graph-based predictive models and methods to assess their performance. At the same time, together with an overview of the field, we provide design guidelines, recommendations, and best practices, as well as an in-depth discussion of open challenges and future research directions.">
@@ -508,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-10-24</td>
           <td>ArXiv</td>
-          <td>6</td>
+          <td>9</td>
           <td>50</td>
         </tr>
 
@@ -520,7 +532,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-01-31</td>
           <td>ArXiv</td>
-          <td>3</td>
+          <td>2</td>
           <td>15</td>
         </tr>
 
@@ -544,7 +556,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-11-10</td>
           <td>ArXiv</td>
-          <td>32</td>
+          <td>49</td>
           <td>6</td>
         </tr>
 
@@ -568,7 +580,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-02-04</td>
           <td>ArXiv</td>
-          <td>4</td>
+          <td>5</td>
           <td>22</td>
         </tr>
 
@@ -580,20 +592,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-01-25</td>
           <td>Inf. Sci.</td>
-          <td>0</td>
           <td>1</td>
-        </tr>
-
-        <tr id="Spatio-temporal forecasting of future values of spatially correlated time series is important across many cyber-physical systems (CPS). Recent studies offer evidence that the use of graph neural networks to capture latent correlations between time series holds a potential for enhanced forecasting. However, most existing methods rely on predefined or self-learning graphs, which are either static or unintentionally dynamic, and thus cannot model the time-varying correlations that exhibit trends and periodicities caused by the regularity of the underlying processes in CPS. To tackle such limitation, we propose Time-aware Graph Structure Learning (TagSL), which extracts time-aware correlations among time series by measuring the interaction of node and time representations in high-dimensional spaces. Notably, we introduce time discrepancy learning that utilizes contrastive learning with distance-based regularization terms to constrain learned spatial correlations to a trend sequence. Additionally, we propose a periodic discriminant function to enable the capture of periodic changes from the state of nodes. Next, we present a Graph Convolution-based Gated Recurrent Unit (GCGRU) that jointly captures spatial and temporal dependencies while learning time-aware and node-specific patterns. Finally, we introduce a unified framework named Time-aware Graph Convolutional Recurrent Network (TGCRN), combining TagSL, and GCGRU in an encoder-decoder architecture for multi-step spatiotemporal forecasting. We report on experiments with TGCRN and popular existing approaches on five real-world datasets, thus providing evidence that TGCRN is capable of advancing the state-of-the-art. We also cover a detailed ablation study and visualization analysis, offering detailed insight into the effectiveness of time-aware structure learning.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2e635d42e768aec7be29fbd203a605c1f011a44d" target='_blank'>Learning Time-Aware Graph Structures for Spatially Correlated Time Series Forecasting</a></td>
-          <td>
-            Minbo Ma, Jilin Hu, Christian S. Jensen, Fei Teng, Peng Han, Zhiqiang Xu, Tian-Jie Li
-          </td>
-          <td>2023-12-27</td>
-          <td>2024 IEEE 40th International Conference on Data Engineering (ICDE)</td>
-          <td>0</td>
-          <td>12</td>
+          <td>1</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419/index.html b/site/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419/index.html
index 7c324925..4747e1ac 100644
--- a/site/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419/index.html
+++ b/site/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:15 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:10 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,7 +484,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-10-24</td>
           <td>ArXiv</td>
-          <td>6</td>
+          <td>9</td>
           <td>50</td>
         </tr>
 
@@ -496,7 +496,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-09-14</td>
           <td>ArXiv</td>
-          <td>30</td>
+          <td>34</td>
           <td>50</td>
         </tr>
 
@@ -508,7 +508,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2022-05-26</td>
           <td>J. Mach. Learn. Res.</td>
-          <td>11</td>
+          <td>14</td>
           <td>50</td>
         </tr>
 
@@ -544,7 +544,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-11-10</td>
           <td>ArXiv</td>
-          <td>32</td>
+          <td>49</td>
           <td>6</td>
         </tr>
 
diff --git a/site/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6/index.html b/site/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6/index.html
index 9f235537..87ffd3db 100644
--- a/site/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6/index.html
+++ b/site/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:31 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:19 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -476,9 +476,9 @@ <h3 data-intro='Recommendations for the article'>
   </thead>
   <tbody>
 
-        <tr id="Foundation models have emerged as a promising approach in time series forecasting (TSF). Existing approaches either fine-tune large language models (LLMs) or build large-scale time-series datasets to develop TSF foundation models. However, these methods face challenges due to the severe cross-domain gap or in-domain heterogeneity. In this paper, we explore a new road to building a TSF foundation model from rich and high-quality natural images, based on the intrinsic similarities between images and time series. To bridge the gap between the two domains, we reformulate the TSF task as an image reconstruction task, which is further processed by a visual masked autoencoder (MAE) self-supervised pre-trained on the ImageNet dataset. Surprisingly, without further adaptation in the time-series domain, the proposed VisionTS could achieve superior zero-shot forecasting performance compared to existing TSF foundation models. With minimal fine-tuning, VisionTS could further improve the forecasting and achieve state-of-the-art performance in most cases. These findings suggest that visual models could be a free lunch for TSF and highlight the potential for future cross-domain research between computer vision and TSF. Our code is publicly available at https://github.com/Keytoyze/VisionTS.">
+        <tr id="Foundation models have emerged as a promising approach in time series forecasting (TSF). Existing approaches either repurpose large language models (LLMs) or build large-scale time series datasets to develop TSF foundation models for universal forecasting. However, these methods face challenges due to the severe cross-domain gap or in-domain heterogeneity. This paper explores a new road to building a TSF foundation model from rich, high-quality natural images. Our key insight is that a visual masked autoencoder, pre-trained on the ImageNet dataset, can naturally be a numeric series forecaster. By reformulating TSF as an image reconstruction task, we bridge the gap between image pre-training and TSF downstream tasks. Surprisingly, without further adaptation in the time-series domain, the proposed VisionTS could achieve superior zero-shot forecasting performance compared to existing TSF foundation models. With fine-tuning for one epoch, VisionTS could further improve the forecasting and achieve state-of-the-art performance in most cases. Extensive experiments reveal intrinsic similarities between images and real-world time series, suggesting visual models may offer a ``free lunch'' for TSF and highlight the potential for future cross-modality research. Our code is publicly available at https://github.com/Keytoyze/VisionTS.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/852d2d0acd8ab45c02a7f5fd921f76c8b5558d99" target='_blank'>VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/7665209f6d696321650c8d7eb1d1635406f44d0d" target='_blank'>VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters</a></td>
           <td>
             Mouxiang Chen, Lefei Shen, Zhuo Li, Xiaoyun Joy Wang, Jianling Sun, Chenghao Liu
           </td>
@@ -488,18 +488,6 @@ <h3 data-intro='Recommendations for the article'>
           <td>3</td>
         </tr>
 
-        <tr id="We present General Time Transformer (GTT), an encoder-only style foundation model for zero-shot multivariate time series forecasting. GTT is pretrained on a large dataset of 200M high-quality time series samples spanning diverse domains. In our proposed framework, the task of multivariate time series forecasting is formulated as a channel-wise next curve shape prediction problem, where each time series sample is represented as a sequence of non-overlapping curve shapes with a unified numerical magnitude. GTT is trained to predict the next curve shape based on a window of past curve shapes in a channel-wise manner. Experimental results demonstrate that GTT exhibits superior zero-shot multivariate forecasting capabilities on unseen time series datasets, even surpassing state-of-the-art supervised baselines. Additionally, we investigate the impact of varying GTT model parameters and training dataset scales, observing that the scaling law also holds in the context of zero-shot multivariate time series forecasting.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/15ea0190593662402d0493958a53d5ca3b8e8406" target='_blank'>Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction</a></td>
-          <td>
-            Cheng Feng, Long Huang, Denis Krompass
-          </td>
-          <td>2024-02-12</td>
-          <td>ArXiv</td>
-          <td>3</td>
-          <td>14</td>
-        </tr>
-
         <tr id="In recent years, generative pre-trained paradigms such as Large Language Models (LLMs) and Large Vision Models (LVMs) have achieved revolutionary advancements and widespread real-world applications. Particularly, the emergence of pre-trained LLMs-based temporal works, compared to previous deep model approaches, has demonstrated superior generalization and robustness, showcasing the potential of generative pre-trained paradigms as foundation models for time series. However, those LLMs-based works mainly focus on cross-modal research, i.e., leveraging the language capabilities of LLMs in time series contexts. Although they have achieved impressive performance, there still exist the issues of concept drift caused by differences in data distribution and inflexibility caused by misalignment of dimensions. To this end, inspired by recent work on LVMs, we reconsider the paradigm of time series modeling. In this paper, we comprehensively explore, for the first time, the effectiveness and superiority of the Generative Pre-trained Diffusion (GPD) paradigm in real-world multivariate time series forecasting (TSF). Specifically, to mitigate performance bias introduced by sophisticated networks, we propose a straightforward MLP diffusion network for unconditional modeling of time series. Then we employ a zero-shot and tuning-free method to predict (generate) future data using historical data as prompts. The GPD paradigm is established on the time series modality, effectively preventing the phenomenon of concept drift, and enabling flexible forecasting of arbitrary lengths. We demonstrate that the GPD paradigm achieves comprehensive performance and generalization comparable to current SOTA LLM-based and deep model paradigms on mainstream benchmarks and various TSF tasks. Extensive experiments validate the potential of the GPD paradigm and its assistance in future related research.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/818793961396bbe4c744932676f0ff9cc8385bed" target='_blank'>Generative Pre-Trained Diffusion Paradigm for Zero-Shot Time Series Forecasting</a></td>
@@ -509,19 +497,19 @@ <h3 data-intro='Recommendations for the article'>
           <td>2024-06-04</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>4</td>
+          <td>5</td>
         </tr>
 
-        <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
+        <tr id="We present General Time Transformer (GTT), an encoder-only style foundation model for zero-shot multivariate time series forecasting. GTT is pretrained on a large dataset of 200M high-quality time series samples spanning diverse domains. In our proposed framework, the task of multivariate time series forecasting is formulated as a channel-wise next curve shape prediction problem, where each time series sample is represented as a sequence of non-overlapping curve shapes with a unified numerical magnitude. GTT is trained to predict the next curve shape based on a window of past curve shapes in a channel-wise manner. Experimental results demonstrate that GTT exhibits superior zero-shot multivariate forecasting capabilities on unseen time series datasets, even surpassing state-of-the-art supervised baselines. Additionally, we investigate the impact of varying GTT model parameters and training dataset scales, observing that the scaling law also holds in the context of zero-shot multivariate time series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4fb78450650894091b0a55b5504a8b0a6f3dec37" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/15ea0190593662402d0493958a53d5ca3b8e8406" target='_blank'>Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction</a></td>
           <td>
-            Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
+            Cheng Feng, Long Huang, Denis Krompass
           </td>
-          <td>2024-03-12</td>
+          <td>2024-02-12</td>
           <td>ArXiv</td>
-          <td>28</td>
-          <td>18</td>
+          <td>4</td>
+          <td>15</td>
         </tr>
 
         <tr id="Over the past years, foundation models have caused a paradigm shift in machine learning due to their unprecedented capabilities for zero-shot and few-shot generalization. However, despite the success of foundation models in modalities such as natural language processing and computer vision, the development of foundation models for time series forecasting has lagged behind. We present Lag-Llama, a general-purpose foundation model for univariate probabilistic time series forecasting based on a decoder-only transformer architecture that uses lags as covariates. Lag-Llama is pretrained on a large corpus of diverse time series data from several domains, and demonstrates strong zero-shot generalization capabilities compared to a wide range of forecasting models on downstream datasets across domains. Moreover, when fine-tuned on relatively small fractions of such previously unseen datasets, Lag-Llama achieves state-of-the-art performance, outperforming prior deep learning approaches, emerging as the best general-purpose model on average. Lag-Llama serves as a strong contender to the current state-of-art in time series forecasting and paves the way for future advancements in foundation models tailored to time series data.">
@@ -532,56 +520,56 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-10-12</td>
           <td>ArXiv</td>
-          <td>16</td>
+          <td>23</td>
           <td>40</td>
         </tr>
 
-        <tr id="Time series data are ubiquitous across various domains, making time series analysis critically important. Traditional time series models are task-specific, featuring singular functionality and limited generalization capacity. Recently, large language foundation models have unveiled their remarkable capabilities for cross-task transferability, zero-shot/few-shot learning, and decision-making explainability. This success has sparked interest in the exploration of foundation models to solve multiple time series challenges simultaneously. There are two main research lines, namely pre-training foundation models from scratch for time series and adapting large language foundation models for time series. They both contribute to the development of a unified model that is highly generalizable, versatile, and comprehensible for time series analysis. This survey offers a 3E analytical framework for comprehensive examination of related research. Specifically, we examine existing works from three dimensions, namely Effectiveness, Efficiency and Explainability. In each dimension, we focus on discussing how related works devise tailored solution by considering unique challenges in the realm of time series. Furthermore, we provide a domain taxonomy to help followers keep up with the domain-specific advancements. In addition, we introduce extensive resources to facilitate the field's development, including datasets, open-source, time series libraries. A GitHub repository is also maintained for resource updates (https://github.com/start2020/Awesome-TimeSeries-LLM-FM).">
+        <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/522828823ff726598675971036fff01b57320578" target='_blank'>A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
           <td>
-            Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung
+            Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
           </td>
-          <td>2024-05-03</td>
+          <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>47</td>
+          <td>69</td>
+          <td>18</td>
         </tr>
 
-        <tr id="In this study, we present aLLM4TS, an innovative framework that adapts Large Language Models (LLMs) for time-series representation learning. Central to our approach is that we reconceive time-series forecasting as a self-supervised, multi-patch prediction task, which, compared to traditional contrastive learning or mask-and-reconstruction methods, captures temporal dynamics in patch representations more effectively. Our strategy encompasses two-stage training: (i). a causal continual pre-training phase on various time-series datasets, anchored on next patch prediction, effectively syncing LLM capabilities with the intricacies of time-series data; (ii). fine-tuning for multi-patch prediction in the targeted time-series context. A distinctive element of our framework is the patch-wise decoding layer, which departs from previous methods reliant on sequence-level decoding. Such a design directly transposes individual patches into temporal sequences, thereby significantly bolstering the model's proficiency in mastering temporal patch-based representations. aLLM4TS demonstrates superior performance in several downstream tasks, proving its effectiveness in deriving temporal representations with enhanced transferability and marking a pivotal advancement in the adaptation of LLMs for time-series analysis.">
+        <tr id="Time Series Forecasting (TSF) is key functionality in numerous fields, including in finance, weather services, and energy management. While TSF methods are emerging these days, many of them require domain-specific data collection and model training and struggle with poor generalization performance on new domains. Foundation models aim to overcome this limitation. Pre-trained on large-scale language or time series data, they exhibit promising inferencing capabilities in new or unseen data. This has spurred a surge in new TSF foundation models. We propose a new benchmark, FoundTS, to enable thorough and fair evaluation and comparison of such models. FoundTS covers a variety of TSF foundation models, including those based on large language models and those pretrained on time series. Next, FoundTS supports different forecasting strategies, including zero-shot, few-shot, and full-shot, thereby facilitating more thorough evaluations. Finally, FoundTS offers a pipeline that standardizes evaluation processes such as dataset splitting, loading, normalization, and few-shot sampling, thereby facilitating fair evaluations. Building on this, we report on an extensive evaluation of TSF foundation models on a broad range of datasets from diverse domains and with different statistical characteristics. Specifically, we identify pros and cons and inherent limitations of existing foundation models, and we identify directions for future model design. We make our code and datasets available at https://anonymous.4open.science/r/FoundTS-C2B0.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/c699624c3cdd5e2ac0d1f67de733bbad1b1c0e37" target='_blank'>Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learning</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/c245a7d561855d7bef6c758e45e772129b52dacb" target='_blank'>FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting</a></td>
           <td>
-            Yuxuan Bian, Xu Ju, Jiangtong Li, Zhijian Xu, Dawei Cheng, Qiang Xu
+            Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang
           </td>
-          <td>2024-02-07</td>
+          <td>2024-10-15</td>
           <td>ArXiv</td>
-          <td>5</td>
-          <td>5</td>
+          <td>0</td>
+          <td>27</td>
         </tr>
 
-        <tr id="Time series has been left behind in the era of pre-training and transfer learning. While research in the fields of natural language processing and computer vision are enjoying progressively larger datasets to train massive models, the most popular time series datasets consist of only tens of thousands of time steps, limiting our ability to study the effectiveness of pre-training and scaling. Recent studies have also cast doubt on the need for expressive models and scale. To alleviate these issues, we introduce three large-scale time series forecasting datasets from the cloud operations (CloudOps) domain, the largest having billions of observations, enabling further study into pre-training and scaling of time series models. We build the empirical groundwork for studying pre-training and scaling of time series models and pave the way for future research by identifying a promising candidate architecture. We show that it is a strong zero-shot baseline and benefits from further scaling, both in model and dataset size. Accompanying these datasets and results is a suite of comprehensive benchmark results comparing classical and deep learning baselines to our pre-trained method - achieving a 27% reduction in error on the largest dataset. Code and datasets can be found https://github.com/SalesforceAIResearch/pretrain-time-series-cloudops.">
+        <tr id="Time series data are ubiquitous across various domains, making time series analysis critically important. Traditional time series models are task-specific, featuring singular functionality and limited generalization capacity. Recently, large language foundation models have unveiled their remarkable capabilities for cross-task transferability, zero-shot/few-shot learning, and decision-making explainability. This success has sparked interest in the exploration of foundation models to solve multiple time series challenges simultaneously. There are two main research lines, namely pre-training foundation models from scratch for time series and adapting large language foundation models for time series. They both contribute to the development of a unified model that is highly generalizable, versatile, and comprehensible for time series analysis. This survey offers a 3E analytical framework for comprehensive examination of related research. Specifically, we examine existing works from three dimensions, namely Effectiveness, Efficiency and Explainability. In each dimension, we focus on discussing how related works devise tailored solution by considering unique challenges in the realm of time series. Furthermore, we provide a domain taxonomy to help followers keep up with the domain-specific advancements. In addition, we introduce extensive resources to facilitate the field's development, including datasets, open-source, time series libraries. A GitHub repository is also maintained for resource updates (https://github.com/start2020/Awesome-TimeSeries-LLM-FM).">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/dbd6463cfff26861f6f3ea8e8d19eff4e6331904" target='_blank'>Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/522828823ff726598675971036fff01b57320578" target='_blank'>A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model</a></td>
           <td>
-            Gerald Woo, Chenghao Liu, Akshat Kumar, Doyen Sahoo
+            Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung
           </td>
-          <td>2023-10-08</td>
+          <td>2024-05-03</td>
           <td>ArXiv</td>
           <td>7</td>
-          <td>22</td>
+          <td>47</td>
         </tr>
 
-        <tr id="Despite the impressive achievements of pre-trained models in the fields of natural language processing (NLP) and computer vision (CV), progress in the domain of time series analysis has been limited. In contrast to NLP and CV, where a single model can handle various tasks, time series analysis still relies heavily on task-specific methods for activities such as classification, anomaly detection, forecasting, and few-shot learning. The primary obstacle to developing a pre-trained model for time series analysis is the scarcity of sufficient training data. In our research, we overcome this obstacle by utilizing pre-trained models from language or CV, which have been trained on billions of data points, and apply them to time series analysis. We assess the effectiveness of the pre-trained transformer model in two ways. Initially, we maintain the original structure of the self-attention and feedforward layers in the residual blocks of the pre-trained language or image model, using the Frozen Pre-trained Transformer (FPT) for time series analysis with the addition of projection matrices for input and output. Additionally, we introduce four unique adapters, designed specifically for downstream tasks based on the pre-trained model, including forecasting and anomaly detection. These adapters are further enhanced with efficient parameter tuning, resulting in superior performance compared to all state-of-the-art methods.Our comprehensive experimental studies reveal that (a) the simple FPT achieves top-tier performance across various time series analysis tasks; and (b) fine-tuning the FPT with the custom-designed adapters can further elevate its performance, outshining specialized task-specific models.">
+        <tr id="In this study, we present aLLM4TS, an innovative framework that adapts Large Language Models (LLMs) for time-series representation learning. Central to our approach is that we reconceive time-series forecasting as a self-supervised, multi-patch prediction task, which, compared to traditional contrastive learning or mask-and-reconstruction methods, captures temporal dynamics in patch representations more effectively. Our strategy encompasses two-stage training: (i). a causal continual pre-training phase on various time-series datasets, anchored on next patch prediction, effectively syncing LLM capabilities with the intricacies of time-series data; (ii). fine-tuning for multi-patch prediction in the targeted time-series context. A distinctive element of our framework is the patch-wise decoding layer, which departs from previous methods reliant on sequence-level decoding. Such a design directly transposes individual patches into temporal sequences, thereby significantly bolstering the model's proficiency in mastering temporal patch-based representations. aLLM4TS demonstrates superior performance in several downstream tasks, proving its effectiveness in deriving temporal representations with enhanced transferability and marking a pivotal advancement in the adaptation of LLMs for time-series analysis.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/15bfe54223d28c74f9e39960f7ff1d03167e285f" target='_blank'>One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/c699624c3cdd5e2ac0d1f67de733bbad1b1c0e37" target='_blank'>Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learning</a></td>
           <td>
-            Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin
+            Yuxuan Bian, Xu Ju, Jiangtong Li, Zhijian Xu, Dawei Cheng, Qiang Xu
           </td>
-          <td>2023-11-24</td>
+          <td>2024-02-07</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>7</td>
+          <td>8</td>
+          <td>8</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8/index.html b/site/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8/index.html
index f28c8fc6..4af81ec8 100644
--- a/site/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8/index.html
+++ b/site/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:12:07 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:53 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,7 +484,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-01-09</td>
           <td>Communications Physics</td>
-          <td>8</td>
+          <td>11</td>
           <td>2</td>
         </tr>
 
@@ -496,8 +496,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3256</td>
-          <td>65</td>
+          <td>3408</td>
+          <td>67</td>
         </tr>
 
         <tr id="Modern societies have an abundance of data yet good system models are rare. Unfortunately, many of the current system identification and machine learning techniques fail to generalize outside of the training set, producing models that violate basic physical laws. This work proposes a novel method for the Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI). SINDy-SI is an iterative method that uses Sum-of-Squares (SOS) programming to learn optimally fitted models while guaranteeing that the learned model satisfies side information, such as symmetries and physical laws. Guided by the principle of Occam's razor, that the simplest or most regularized best fitted model is typically the superior choice, during each iteration SINDy-SI prunes the basis functions associated with small coefficients, yielding a sparse dynamical model upon termination. Through several numerical experiments we will show how the combination of side information constraints and sparse polynomial representation cultivates dynamical models that obey known physical laws while displaying impressive generalized performance beyond the training set.">
@@ -520,8 +520,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>111</td>
-          <td>65</td>
+          <td>119</td>
+          <td>67</td>
         </tr>
 
         <tr id="Sparse regression has emerged as a popular technique for learning dynamical systems from temporal data, beginning with the SINDy (Sparse Identification of Nonlinear Dynamics) framework proposed by arXiv:1509.03580. Quantifying the uncertainty inherent in differential equations learned from data remains an open problem, thus we propose leveraging recent advances in statistical inference for sparse regression to address this issue. Focusing on systems of ordinary differential equations (ODEs), SINDy assumes that each equation is a parsimonious linear combination of a few candidate functions, such as polynomials, and uses methods such as sequentially-thresholded least squares or the Lasso to identify a small subset of these functions that govern the system's dynamics. We instead employ bias-corrected versions of the Lasso and ridge regression estimators, as well as an empirical Bayes variable selection technique known as SEMMS, to estimate each ODE as a linear combination of terms that are statistically significant. We demonstrate through simulations that this approach allows us to recover the functional terms that correctly describe the dynamics more often than existing methods that do not account for uncertainty.">
@@ -544,8 +544,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>250</td>
-          <td>12</td>
+          <td>289</td>
+          <td>13</td>
         </tr>
 
         <tr id="PySINDy is a Python package for the discovery of governing dynamical systems models from data. In particular, PySINDy provides tools for applying the sparse identification of nonlinear dynamics (SINDy) (Brunton et al. 2016) approach to model discovery. In this work we provide a brief description of the mathematical underpinnings of SINDy, an overview and demonstration of the features implemented in PySINDy (with code examples), practical advice for users, and a list of potential extensions to PySINDy. Software is available at this https URL.">
@@ -556,8 +556,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-04-17</td>
           <td>arXiv: Dynamical Systems</td>
-          <td>49</td>
-          <td>65</td>
+          <td>51</td>
+          <td>67</td>
         </tr>
 
         <tr id="Data-driven modeling frameworks that adopt sparse regression techniques, such as sparse identification of nonlinear dynamics (SINDy) and its modifications, are developed to resolve difficulties in extracting underlying dynamics from experimental data. In contrast to neural-network-based methods, these methods are designed to obtain white-box analytical models. In this work, we incorporate the concept of SINDy and knowledge in the field of classical mechanics to identify interpretable and sparse expressions of total energy and the Lagrangian that shelters the hidden dynamics. Moreover, our method (hereafter referred as Lagrangian-SINDy) is developed to use knowledge of simple systems that form the system being analyzed to ensure the likelihood of correct results and to improve the learning pace. Lagrangian-SINDy is highly accurate in discovering interpretable dynamics via energy-related physical quantities. Its performance is validated with three popular multi-DOF nonlinear dynamical systems, namely the spherical pendulum, double pendulum and cart-pendulum system. Comparisons with other SINDy-based methods are made and Lagrangian-SINDy is found to provide the most compact analytical models.">
@@ -568,8 +568,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2020-01-31</td>
           <td>IEEE Robotics and Automation Letters</td>
-          <td>23</td>
-          <td>23</td>
+          <td>24</td>
+          <td>24</td>
         </tr>
 
         <tr id="With the rapid increase of available data for complex systems, there is great interest in the extraction of physically relevant information from massive datasets. Recently, a framework called Sparse Identification of Nonlinear Dynamics (SINDy) has been introduced to identify the governing equations of dynamical systems from simulation data. In this study, we extend SINDy to stochastic dynamical systems which are frequently used to model biophysical processes. We prove the asymptotic correctness of stochastic SINDy in the infinite data limit, both in the original and projected variables. We discuss algorithms to solve the sparse regression problem arising from the practical implementation of SINDy and show that cross validation is an essential tool to determine the right level of sparsity. We demonstrate the proposed methodology on two test systems, namely, the diffusion in a one-dimensional potential and the projected dynamics of a two-dimensional diffusion process.">
@@ -580,20 +580,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2017-12-06</td>
           <td>The Journal of chemical physics</td>
-          <td>193</td>
-          <td>45</td>
-        </tr>
-
-        <tr id="Discovering the partial differential equations underlying spatio-temporal datasets from very limited and highly noisy observations is of paramount interest in many scientific fields. However, it remains an open question to know when model discovery algorithms based on sparse regression can actually recover the underlying physical processes. In this work, we show the design matrices used to infer the equations by sparse regression can violate the irrepresentability condition (IRC) of the Lasso, even when derived from analytical PDE solutions (i.e. without additional noise). Sparse regression techniques which can recover the true underlying model under violated IRC conditions are therefore required, leading to the introduction of the randomised adaptive Lasso. We show once the latter is integrated within the deep learning model discovery framework DeepMod, a wide variety of nonlinear and chaotic canonical PDEs can be recovered: (1) up to $\mathcal{O}(2)$ higher noise-to-sample ratios than state-of-the-art algorithms, (2) with a single set of hyperparameters, which paves the road towards truly automated model discovery.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/45633bde87d3fecdb0ef577036f2465e70c51f24" target='_blank'>Sparsistent Model Discovery</a></td>
-          <td>
-            Georges Tod, G. Both, R. Kusters
-          </td>
-          <td>2021-06-22</td>
-          <td>ArXiv</td>
-          <td>1</td>
-          <td>12</td>
+          <td>200</td>
+          <td>46</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f/index.html b/site/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f/index.html
index 508bbb00..a18e2cf9 100644
--- a/site/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f/index.html
+++ b/site/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:18 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:15 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,104 +484,104 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-10-12</td>
           <td>ArXiv</td>
-          <td>16</td>
+          <td>23</td>
           <td>40</td>
         </tr>
 
-        <tr id="Large language models (LLMs) are being applied to time series tasks, particularly time series forecasting. However, are language models actually useful for time series? After a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade the forecasting results -- in most cases the results even improved. We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and reveal that patching and attention structures perform similarly to state-of-the-art LLM-based forecasters.">
+        <tr id="Motivated by the recent success of time-series foundation models for zero-shot forecasting, we present a methodology for $\textit{in-context fine-tuning}$ of a time-series foundation model. In particular, we design a pretrained foundation model that can be prompted (at inference time) with multiple time-series examples, in order to forecast a target time-series into the future. Our foundation model is specifically trained to utilize examples from multiple related time-series in its context window (in addition to the history of the target time-series) to help it adapt to the specific distribution of the target domain at inference time. We show that such a foundation model that uses in-context examples at inference time can obtain much better performance on popular forecasting benchmarks compared to supervised deep learning methods, statistical models, as well as other time-series foundation models. Interestingly, our in-context fine-tuning approach even rivals the performance of a foundation model that is explicitly fine-tuned on the target domain.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/df0d604b8e8e3b2947d9865d735f204c08635012" target='_blank'>Are Language Models Actually Useful for Time Series Forecasting?</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/1216a29c0beb36a2b621b59d36877edf05afcf8c" target='_blank'>In-Context Fine-Tuning for Time-Series Foundation Models</a></td>
           <td>
-            Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen
+            Abhimanyu Das, Matthew Faw, Rajat Sen, Yichen Zhou
           </td>
-          <td>2024-06-22</td>
+          <td>2024-10-31</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>1</td>
+          <td>0</td>
+          <td>14</td>
         </tr>
 
-        <tr id="Time series forecasting holds significant importance in many real-world dynamic systems and has been extensively studied. Unlike natural language process (NLP) and computer vision (CV), where a single large model can tackle multiple tasks, models for time series forecasting are often specialized, necessitating distinct designs for different tasks and applications. While pre-trained foundation models have made impressive strides in NLP and CV, their development in time series domains has been constrained by data sparsity. Recent studies have revealed that large language models (LLMs) possess robust pattern recognition and reasoning abilities over complex sequences of tokens. However, the challenge remains in effectively aligning the modalities of time series data and natural language to leverage these capabilities. In this work, we present Time-LLM, a reprogramming framework to repurpose LLMs for general time series forecasting with the backbone language models kept intact. We begin by reprogramming the input time series with text prototypes before feeding it into the frozen LLM to align the two modalities. To augment the LLM's ability to reason with time series data, we propose Prompt-as-Prefix (PaP), which enriches the input context and directs the transformation of reprogrammed input patches. The transformed time series patches from the LLM are finally projected to obtain the forecasts. Our comprehensive evaluations demonstrate that Time-LLM is a powerful time series learner that outperforms state-of-the-art, specialized forecasting models. Moreover, Time-LLM excels in both few-shot and zero-shot learning scenarios.">
+        <tr id="Large language models (LLMs) are being applied to time series forecasting. But are language models actually useful for time series? In a series of ablation studies on three recent and popular LLM-based time series forecasting methods, we find that removing the LLM component or replacing it with a basic attention layer does not degrade forecasting performance -- in most cases, the results even improve! We also find that despite their significant computational cost, pretrained LLMs do no better than models trained from scratch, do not represent the sequential dependencies in time series, and do not assist in few-shot settings. Additionally, we explore time series encoders and find that patching and attention structures perform similarly to LLM-based forecasters.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277" target='_blank'>Time-LLM: Time Series Forecasting by Reprogramming Large Language Models</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/df0d604b8e8e3b2947d9865d735f204c08635012" target='_blank'>Are Language Models Actually Useful for Time Series Forecasting?</a></td>
           <td>
-            Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen
+            Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen
           </td>
-          <td>2023-10-03</td>
+          <td>2024-06-22</td>
           <td>ArXiv</td>
-          <td>134</td>
-          <td>9</td>
+          <td>13</td>
+          <td>3</td>
         </tr>
 
-        <tr id="Recent Transformer-based large language models (LLMs) demonstrate in-context learning ability to perform various functions based solely on the provided context, without updating model parameters. To fully utilize the in-context capabilities in time series forecasting (TSF) problems, unlike previous Transformer-based or LLM-based time series forecasting methods, we reformulate"time series forecasting tasks"as input tokens by constructing a series of (lookback, future) pairs within the tokens. This method aligns more closely with the inherent in-context mechanisms, and is more parameter-efficient without the need of using pre-trained LLM parameters. Furthermore, it addresses issues such as overfitting in existing Transformer-based TSF models, consistently achieving better performance across full-data, few-shot, and zero-shot settings compared to previous architectures.">
+        <tr id="Inspired by recent advancements in large language models (LLMs) for Natural Language Processing (NLP), there has been a surge in research focused on developing foundational models for time series forecasting. One approach involves training LLM architectures on tokenized time series data using cross-entropy loss. Although this method has demonstrated promising results, cross-entropy loss is primarily designed for classification tasks and does not account for the distance between classes. To address this limitation, we propose using the Wasserstein loss for such architectures. To validate our approach, we fine-tuned a foundational time series model on $22$ zero-shot datasets, comparing the performance of cross-entropy loss with that of Wasserstein loss. Our results demonstrate that replacing cross-entropy loss with Wasserstein loss significantly improves point estimation.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/3777e8d6cda553cbd9e3bbd6484e613386f3a59e" target='_blank'>In-context Time Series Predictor</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/341fc8f00805b53471ae91bc2f7e1d5b4d281048" target='_blank'>Fine-Tuning a Time Series Foundation Model with Wasserstein Loss</a></td>
           <td>
-            Jiecheng Lu, Yan Sun, Shihao Yang
+            Andrei Chernov
           </td>
-          <td>2024-05-23</td>
+          <td>2024-09-18</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>1</td>
+          <td>0</td>
         </tr>
 
-        <tr id="Multivariate time-series forecasting is vital in various domains, e.g., economic planning and weather prediction. Deep train-from-scratch models have exhibited effective performance yet require large amounts of data, which limits real-world applicability. Recently, researchers have leveraged the representation learning transferability of pre-trained Large Language Models (LLMs) to handle limited non-linguistic datasets effectively. However, incorporating LLMs with time-series data presents challenges of limited adaptation due to different compositions between time-series and linguistic data, and the inability to process multi-scale temporal information. To tackle these challenges, we propose LLM4TS, a framework for time-series forecasting with pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the \textit{time-series alignment} stage to align LLMs with the nuances of time-series data, and the \textit{forecasting fine-tuning} stage for downstream time-series forecasting tasks. Furthermore, our framework features a novel two-level aggregation method that integrates multi-scale temporal data within pre-trained LLMs, enhancing their ability to interpret time-specific information. In experiments across 7 time-series forecasting datasets, LLM4TS is superior to existing state-of-the-art methods compared with trained-from-scratch models in full-shot scenarios, and also achieves an average improvement of 6.84% in MSE in few-shot scenarios. In addition, evaluations compared with different self-supervised learning approaches highlight LLM4TS's effectiveness with representation learning in forecasting tasks.">
+        <tr id="Time series forecasting holds significant importance in many real-world dynamic systems and has been extensively studied. Unlike natural language process (NLP) and computer vision (CV), where a single large model can tackle multiple tasks, models for time series forecasting are often specialized, necessitating distinct designs for different tasks and applications. While pre-trained foundation models have made impressive strides in NLP and CV, their development in time series domains has been constrained by data sparsity. Recent studies have revealed that large language models (LLMs) possess robust pattern recognition and reasoning abilities over complex sequences of tokens. However, the challenge remains in effectively aligning the modalities of time series data and natural language to leverage these capabilities. In this work, we present Time-LLM, a reprogramming framework to repurpose LLMs for general time series forecasting with the backbone language models kept intact. We begin by reprogramming the input time series with text prototypes before feeding it into the frozen LLM to align the two modalities. To augment the LLM's ability to reason with time series data, we propose Prompt-as-Prefix (PaP), which enriches the input context and directs the transformation of reprogrammed input patches. The transformed time series patches from the LLM are finally projected to obtain the forecasts. Our comprehensive evaluations demonstrate that Time-LLM is a powerful time series learner that outperforms state-of-the-art, specialized forecasting models. Moreover, Time-LLM excels in both few-shot and zero-shot learning scenarios.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/663dc434fabc1cf1d3e85fff3f7ddcd313035d18" target='_blank'>LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277" target='_blank'>Time-LLM: Time Series Forecasting by Reprogramming Large Language Models</a></td>
           <td>
-            Ching Chang, Wenjie Peng, Tien-Fu Chen
+            Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen
           </td>
-          <td>2023-08-16</td>
+          <td>2023-10-03</td>
           <td>ArXiv</td>
-          <td>16</td>
-          <td>2</td>
+          <td>195</td>
+          <td>9</td>
         </tr>
 
-        <tr id="Time series data are ubiquitous across various domains, making time series analysis critically important. Traditional time series models are task-specific, featuring singular functionality and limited generalization capacity. Recently, large language foundation models have unveiled their remarkable capabilities for cross-task transferability, zero-shot/few-shot learning, and decision-making explainability. This success has sparked interest in the exploration of foundation models to solve multiple time series challenges simultaneously. There are two main research lines, namely pre-training foundation models from scratch for time series and adapting large language foundation models for time series. They both contribute to the development of a unified model that is highly generalizable, versatile, and comprehensible for time series analysis. This survey offers a 3E analytical framework for comprehensive examination of related research. Specifically, we examine existing works from three dimensions, namely Effectiveness, Efficiency and Explainability. In each dimension, we focus on discussing how related works devise tailored solution by considering unique challenges in the realm of time series. Furthermore, we provide a domain taxonomy to help followers keep up with the domain-specific advancements. In addition, we introduce extensive resources to facilitate the field's development, including datasets, open-source, time series libraries. A GitHub repository is also maintained for resource updates (https://github.com/start2020/Awesome-TimeSeries-LLM-FM).">
+        <tr id="Recent Transformer-based large language models (LLMs) demonstrate in-context learning ability to perform various functions based solely on the provided context, without updating model parameters. To fully utilize the in-context capabilities in time series forecasting (TSF) problems, unlike previous Transformer-based or LLM-based time series forecasting methods, we reformulate"time series forecasting tasks"as input tokens by constructing a series of (lookback, future) pairs within the tokens. This method aligns more closely with the inherent in-context mechanisms, and is more parameter-efficient without the need of using pre-trained LLM parameters. Furthermore, it addresses issues such as overfitting in existing Transformer-based TSF models, consistently achieving better performance across full-data, few-shot, and zero-shot settings compared to previous architectures.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/522828823ff726598675971036fff01b57320578" target='_blank'>A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/3777e8d6cda553cbd9e3bbd6484e613386f3a59e" target='_blank'>In-context Time Series Predictor</a></td>
           <td>
-            Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung
+            Jiecheng Lu, Yan Sun, Shihao Yang
           </td>
-          <td>2024-05-03</td>
+          <td>2024-05-23</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>47</td>
+          <td>2</td>
+          <td>2</td>
         </tr>
 
-        <tr id="Time-series forecasting (TSF) finds broad applications in real-world scenarios. Prompting off-the-shelf Large Language Models (LLMs) demonstrates strong zero-shot TSF capabilities while preserving computational efficiency. However, existing prompting methods oversimplify TSF as language next-token predictions, overlooking its dynamic nature and lack of integration with state-of-the-art prompt strategies such as Chain-of-Thought. Thus, we propose LSTPrompt, a novel approach for prompting LLMs in zero-shot TSF tasks. LSTPrompt decomposes TSF into short-term and long-term forecasting sub-tasks, tailoring prompts to each. LSTPrompt guides LLMs to regularly reassess forecasting mechanisms to enhance adaptability. Extensive evaluations demonstrate consistently better performance of LSTPrompt than existing prompting methods, and competitive results compared to foundation TSF models.">
+        <tr id="Time Series Forecasting (TSF) is key functionality in numerous fields, including in finance, weather services, and energy management. While TSF methods are emerging these days, many of them require domain-specific data collection and model training and struggle with poor generalization performance on new domains. Foundation models aim to overcome this limitation. Pre-trained on large-scale language or time series data, they exhibit promising inferencing capabilities in new or unseen data. This has spurred a surge in new TSF foundation models. We propose a new benchmark, FoundTS, to enable thorough and fair evaluation and comparison of such models. FoundTS covers a variety of TSF foundation models, including those based on large language models and those pretrained on time series. Next, FoundTS supports different forecasting strategies, including zero-shot, few-shot, and full-shot, thereby facilitating more thorough evaluations. Finally, FoundTS offers a pipeline that standardizes evaluation processes such as dataset splitting, loading, normalization, and few-shot sampling, thereby facilitating fair evaluations. Building on this, we report on an extensive evaluation of TSF foundation models on a broad range of datasets from diverse domains and with different statistical characteristics. Specifically, we identify pros and cons and inherent limitations of existing foundation models, and we identify directions for future model design. We make our code and datasets available at https://anonymous.4open.science/r/FoundTS-C2B0.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/bbf272d92b76cc5f1e6f793dcfeb5edd4abed7ef" target='_blank'>LSTPrompt: Large Language Models as Zero-Shot Time Series Forecasters by Long-Short-Term Prompting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/c245a7d561855d7bef6c758e45e772129b52dacb" target='_blank'>FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting</a></td>
           <td>
-            Haoxin Liu, Zhiyuan Zhao, Jindong Wang, Harshavardhan Kamarthi, B. A. Prakash
+            Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang
           </td>
-          <td>2024-02-25</td>
+          <td>2024-10-15</td>
           <td>ArXiv</td>
-          <td>6</td>
-          <td>7</td>
+          <td>0</td>
+          <td>27</td>
         </tr>
 
-        <tr id="Unlike natural language processing and computer vision, the development of Foundation Models (FMs) for time series forecasting is blocked due to data scarcity. While recent efforts are focused on building such FMs by unlocking the potential of language models (LMs) for time series analysis, dedicated parameters for various downstream forecasting tasks need training, which hinders the common knowledge sharing across domains. Moreover, data owners may hesitate to share the access to local data due to privacy concerns and copyright protection, which makes it impossible to simply construct a FM on cross-domain training instances. To address these issues, we propose Time-FFM, a Federated Foundation Model for Time series forecasting by leveraging pretrained LMs. Specifically, we begin by transforming time series into the modality of text tokens. To bootstrap LMs for time series reasoning, we propose a prompt adaption module to determine domain-customized prompts dynamically instead of artificially. Given the data heterogeneity across domains, we design a personalized federated training strategy by learning global encoders and local prediction heads. Our comprehensive experiments indicate that Time-FFM outperforms state-of-the-arts and promises effective few-shot and zero-shot forecaster.">
+        <tr id="Multivariate time-series forecasting is vital in various domains, e.g., economic planning and weather prediction. Deep train-from-scratch models have exhibited effective performance yet require large amounts of data, which limits real-world applicability. Recently, researchers have leveraged the representation learning transferability of pre-trained Large Language Models (LLMs) to handle limited non-linguistic datasets effectively. However, incorporating LLMs with time-series data presents challenges of limited adaptation due to different compositions between time-series and linguistic data, and the inability to process multi-scale temporal information. To tackle these challenges, we propose LLM4TS, a framework for time-series forecasting with pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the \textit{time-series alignment} stage to align LLMs with the nuances of time-series data, and the \textit{forecasting fine-tuning} stage for downstream time-series forecasting tasks. Furthermore, our framework features a novel two-level aggregation method that integrates multi-scale temporal data within pre-trained LLMs, enhancing their ability to interpret time-specific information. In experiments across 7 time-series forecasting datasets, LLM4TS is superior to existing state-of-the-art methods compared with trained-from-scratch models in full-shot scenarios, and also achieves an average improvement of 6.84% in MSE in few-shot scenarios. In addition, evaluations compared with different self-supervised learning approaches highlight LLM4TS's effectiveness with representation learning in forecasting tasks.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/68a65925ac6d270ca27a4fe99e58cf2ed0795821" target='_blank'>Time-FFM: Towards LM-Empowered Federated Foundation Model for Time Series Forecasting</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/663dc434fabc1cf1d3e85fff3f7ddcd313035d18" target='_blank'>LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters</a></td>
           <td>
-            Qingxiang Liu, Xu Liu, Chenghao Liu, Qingsong Wen, Yuxuan Liang
+            Ching Chang, Wenjie Peng, Tien-Fu Chen
           </td>
-          <td>2024-05-23</td>
+          <td>2023-08-16</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>6</td>
+          <td>22</td>
+          <td>3</td>
         </tr>
 
-        <tr id="Despite the impressive achievements of pre-trained models in the fields of natural language processing (NLP) and computer vision (CV), progress in the domain of time series analysis has been limited. In contrast to NLP and CV, where a single model can handle various tasks, time series analysis still relies heavily on task-specific methods for activities such as classification, anomaly detection, forecasting, and few-shot learning. The primary obstacle to developing a pre-trained model for time series analysis is the scarcity of sufficient training data. In our research, we overcome this obstacle by utilizing pre-trained models from language or CV, which have been trained on billions of data points, and apply them to time series analysis. We assess the effectiveness of the pre-trained transformer model in two ways. Initially, we maintain the original structure of the self-attention and feedforward layers in the residual blocks of the pre-trained language or image model, using the Frozen Pre-trained Transformer (FPT) for time series analysis with the addition of projection matrices for input and output. Additionally, we introduce four unique adapters, designed specifically for downstream tasks based on the pre-trained model, including forecasting and anomaly detection. These adapters are further enhanced with efficient parameter tuning, resulting in superior performance compared to all state-of-the-art methods.Our comprehensive experimental studies reveal that (a) the simple FPT achieves top-tier performance across various time series analysis tasks; and (b) fine-tuning the FPT with the custom-designed adapters can further elevate its performance, outshining specialized task-specific models.">
+        <tr id="Deep learning for time series forecasting has seen significant advancements over the past decades. However, despite the success of large-scale pre-training in language and vision domains, pre-trained time series models remain limited in scale and operate at a high cost, hindering the development of larger capable forecasting models in real-world applications. In response, we introduce Time-MoE, a scalable and unified architecture designed to pre-train larger, more capable forecasting foundation models while reducing inference costs. By leveraging a sparse mixture-of-experts (MoE) design, Time-MoE enhances computational efficiency by activating only a subset of networks for each prediction, reducing computational load while maintaining high model capacity. This allows Time-MoE to scale effectively without a corresponding increase in inference costs. Time-MoE comprises a family of decoder-only transformer models that operate in an auto-regressive manner and support flexible forecasting horizons with varying input context lengths. We pre-trained these models on our newly introduced large-scale data Time-300B, which spans over 9 domains and encompassing over 300 billion time points. For the first time, we scaled a time series foundation model up to 2.4 billion parameters, achieving significantly improved forecasting precision. Our results validate the applicability of scaling laws for training tokens and model size in the context of time series forecasting. Compared to dense models with the same number of activated parameters or equivalent computation budgets, our models consistently outperform them by large margin. These advancements position Time-MoE as a state-of-the-art solution for tackling real-world time series forecasting challenges with superior capability, efficiency, and flexibility.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/15bfe54223d28c74f9e39960f7ff1d03167e285f" target='_blank'>One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/2cb3044ef42c7ee022a988864028b80ce977072c" target='_blank'>Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts</a></td>
           <td>
-            Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin
+            X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin
           </td>
-          <td>2023-11-24</td>
+          <td>2024-09-24</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>7</td>
+          <td>2</td>
+          <td>9</td>
         </tr>
 
   </tbody>
diff --git a/site/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90/index.html b/site/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90/index.html
index 9ba0042a..c5ccad58 100644
--- a/site/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90/index.html
+++ b/site/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90/index.html
@@ -445,7 +445,7 @@ <h1 id="_1"></h1>
 
 <body>
   <p>
-  <i class="footer">This page was last updated on 2024-09-11 16:11:34 UTC</i>
+  <i class="footer">This page was last updated on 2024-11-11 06:05:30 UTC</i>
   </p>
 
   <div class="note info" onclick="startIntro()">
@@ -484,8 +484,20 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>4</td>
-          <td>66</td>
+          <td>21</td>
+          <td>67</td>
+        </tr>
+
+        <tr id="In natural language processing and computer vision, self-supervised pre-training on large datasets unlocks foundational model capabilities across domains and tasks. However, this potential has not yet been realised in time series analysis, where existing methods disregard the heterogeneous nature of time series characteristics. Time series are prevalent in many domains, including medicine, engineering, natural sciences, and finance, but their characteristics vary significantly in terms of variate count, inter-variate relationships, temporal dynamics, and sampling frequency. This inherent heterogeneity across domains prevents effective pre-training on large time series corpora. To address this issue, we introduce OTiS, an open model for general time series analysis, that has been specifically designed to handle multi-domain heterogeneity. We propose a novel pre-training paradigm including a tokeniser with learnable domain-specific signatures, a dual masking strategy to capture temporal causality, and a normalised cross-correlation loss to model long-range dependencies. Our model is pre-trained on a large corpus of 640,187 samples and 11 billion time points spanning 8 distinct domains, enabling it to analyse time series from any (unseen) domain. In comprehensive experiments across 15 diverse applications - including classification, regression, and forecasting - OTiS showcases its ability to accurately capture domain-specific data characteristics and demonstrates its competitiveness against state-of-the-art baselines. Our code and pre-trained weights are publicly available at https://github.com/oetu/otis.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/67381675766ffe648e1346cb19d0d7318366bbbe" target='_blank'>Towards Generalisable Time Series Understanding Across Domains</a></td>
+          <td>
+            Ozgun Turgut, Philip Muller, M. Menten, Daniel Rueckert
+          </td>
+          <td>2024-10-09</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>15</td>
         </tr>
 
         <tr id="Large pre-trained models have been vital in recent advancements in domains like language and vision, making model training for individual downstream tasks more efficient and provide superior performance. However, tackling time-series analysis tasks usually involves designing and training a separate model from scratch leveraging training data and domain expertise specific to the task. We tackle a significant challenge for pre-training a foundational time-series model from multi-domain time-series datasets: extracting semantically useful tokenized inputs to the model across heterogenous time-series from different domains. We propose Large Pre-trained Time-series Models (LPTM) that introduces a novel method of \textit{adaptive segmentation} that automatically identifies optimal dataset-specific segmentation strategy during pre-training. This enables LPTM to perform similar to or better than domain-specific state-of-art model when fine-tuned to different downstream time-series analysis tasks and under zero-shot settings. LPTM achieves superior forecasting and time-series classification results taking up to 40% less data and 50% less training time compared to state-of-art baselines.">
@@ -496,8 +508,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2023-11-19</td>
           <td>ArXiv</td>
-          <td>2</td>
-          <td>7</td>
+          <td>4</td>
+          <td>8</td>
         </tr>
 
         <tr id="The field of general time series analysis has recently begun to explore unified modeling, where a common architectural backbone can be retrained on a specific task for a specific dataset. In this work, we approach unification from a complementary vantage point: unification across tasks and domains. To this end, we explore the impact of discrete, learnt, time series data representations that enable generalist, cross-domain training. Our method, TOTEM, or TOkenized Time Series EMbeddings, proposes a simple tokenizer architecture that embeds time series data from varying domains using a discrete vectorized representation learned in a self-supervised manner. TOTEM works across multiple tasks and domains with minimal to no tuning. We study the efficacy of TOTEM with an extensive evaluation on 17 real world time series datasets across 3 tasks. We evaluate both the specialist (i.e., training a model on each domain) and generalist (i.e., training a single model on many domains) settings, and show that TOTEM matches or outperforms previous best methods on several popular benchmarks. The code can be found at: https://github.com/SaberaTalukder/TOTEM.">
@@ -508,7 +520,7 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-26</td>
           <td>ArXiv</td>
-          <td>4</td>
+          <td>7</td>
           <td>4</td>
         </tr>
 
@@ -520,8 +532,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-01-08</td>
           <td>ArXiv</td>
-          <td>5</td>
           <td>6</td>
+          <td>7</td>
         </tr>
 
         <tr id="Despite the impressive achievements of pre-trained models in the fields of natural language processing (NLP) and computer vision (CV), progress in the domain of time series analysis has been limited. In contrast to NLP and CV, where a single model can handle various tasks, time series analysis still relies heavily on task-specific methods for activities such as classification, anomaly detection, forecasting, and few-shot learning. The primary obstacle to developing a pre-trained model for time series analysis is the scarcity of sufficient training data. In our research, we overcome this obstacle by utilizing pre-trained models from language or CV, which have been trained on billions of data points, and apply them to time series analysis. We assess the effectiveness of the pre-trained transformer model in two ways. Initially, we maintain the original structure of the self-attention and feedforward layers in the residual blocks of the pre-trained language or image model, using the Frozen Pre-trained Transformer (FPT) for time series analysis with the addition of projection matrices for input and output. Additionally, we introduce four unique adapters, designed specifically for downstream tasks based on the pre-trained model, including forecasting and anomaly detection. These adapters are further enhanced with efficient parameter tuning, resulting in superior performance compared to all state-of-the-art methods.Our comprehensive experimental studies reveal that (a) the simple FPT achieves top-tier performance across various time series analysis tasks; and (b) fine-tuning the FPT with the custom-designed adapters can further elevate its performance, outshining specialized task-specific models.">
@@ -533,18 +545,18 @@ <h3 data-intro='Recommendations for the article'>
           <td>2023-11-24</td>
           <td>ArXiv</td>
           <td>4</td>
-          <td>7</td>
+          <td>8</td>
         </tr>
 
         <tr id="We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models. Chronos tokenizes time series values using scaling and quantization into a fixed vocabulary and trains existing transformer-based language model architectures on these tokenized time series via the cross-entropy loss. We pretrained Chronos models based on the T5 family (ranging from 20M to 710M parameters) on a large collection of publicly available datasets, complemented by a synthetic dataset that we generated via Gaussian processes to improve generalization. In a comprehensive benchmark consisting of 42 datasets, and comprising both classical local models and deep learning methods, we show that Chronos models: (a) significantly outperform other methods on datasets that were part of the training corpus; and (b) have comparable and occasionally superior zero-shot performance on new datasets, relative to methods that were trained specifically on them. Our results demonstrate that Chronos models can leverage time series data from diverse domains to improve zero-shot accuracy on unseen forecasting tasks, positioning pretrained models as a viable tool to greatly simplify forecasting pipelines.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4fb78450650894091b0a55b5504a8b0a6f3dec37" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
           <td>
             Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>28</td>
+          <td>69</td>
           <td>18</td>
         </tr>
 
@@ -556,8 +568,8 @@ <h3 data-intro='Recommendations for the article'>
           </td>
           <td>2024-02-04</td>
           <td>ArXiv</td>
-          <td>32</td>
-          <td>22</td>
+          <td>71</td>
+          <td>25</td>
         </tr>
 
   </tbody>
diff --git a/site/search/search_index.json b/site/search/search_index.json
index 6ac03311..093b2abb 100644
--- a/site/search/search_index.json
+++ b/site/search/search_index.json
@@ -1 +1 @@
-{"config":{"lang":["en"],"separator":"[\\s\\-]+","pipeline":["stopWordFilter"]},"docs":[{"location":"","title":"Home","text":"Introduction <p>Welcome to Team VPE's Literature Survey System! This project leverages the powerful Semantic Scholar's Recommendation API to provide you with highly relevant research article recommendations based on your curated lists of articles.</p> <p>Moreover, this website contains curated collection of references on differentiable and learned simulator algorithms for developing digital twins with applications in drug discovery and development.</p> Features <ul> <li>List-based recommendations: For each topic, we use a manually curated list of positive articles. We then select an equal number of negative articles from other topics. Semantics Scholar's recommendation system uses the set of positive and negative to provide up to 300 recommendations per topic.</li> <li>Single-paper recommendations: For each article in our curated lists, Semantics Scholar's recommendation system returns up to 10 related articles.</li> </ul> How it works? <ul> <li>Scroll to the top of this page and click one of the tabs to open the relevant section.</li> <li>Each topic contains two tables. The first table lists the manually curated articles, and the second table lists the articles recommended for that topic by the recommendation system. The last column of the first table contains a hyperlink that opens a new tab with articles recommended for that specific article.</li> <li>Click on the help icon at the top of the page to learn more about that section.</li> </ul>         Check out our GitHub open_in_new repository to learn how to build one for yourself."},{"location":"Symbolic%20regression/","title":"Symbolic regression","text":"<p> This page was last updated on 2024-09-11 16:12:18 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p> </p> <p> </p>"},{"location":"Symbolic%20regression/#manually_curated_articles","title":"Manually curated articles on Symbolic regression","text":"Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index                View recommendations                visibility_off                  Discovering governing equations from data by sparse identification of nonlinear dynamical systems                               S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences, Proceedings of the National Academy of Sciences of the United States of America 3256 65 open_in_new visibility_off                  Robust learning from noisy, incomplete, high-dimensional experimental data via physically constrained symbolic regression                               Patrick A. K. Reinbold, Logan Kageorge, M. Schatz, R. Grigoriev            2021-02-24 Nature Communications 86 23 open_in_new visibility_off                  Data-driven discovery of coordinates and governing equations                               Kathleen P. Champion, Bethany Lusch, J. Kutz, S. Brunton            2019-03-29 Proceedings of the National Academy of Sciences of the United States of America 615 65 open_in_new visibility_off                  Chaos as an intermittently forced linear system                               S. Brunton, Bingni W. Brunton, J. Proctor, E. Kaiser, J. Kutz            2016-08-18 Nature Communications 451 65 open_in_new visibility_off                  Sparse identification of nonlinear dynamics for model predictive control in the low-data limit                               E. Kaiser, J. Kutz, S. Brunton            2017-11-15 Proceedings. Mathematical, Physical, and Engineering Sciences, Proceedings of the Royal Society A 435 65 open_in_new visibility_off                  Inferring Biological Networks by Sparse Identification of Nonlinear Dynamics                               N. Mangan, S. Brunton, J. Proctor, J. Kutz            2016-05-26 IEEE Transactions on Molecular Biological and Multi-Scale Communications, IEEE Transactions on Molecular, Biological and Multi-Scale Communications 321 65 open_in_new visibility_off                  SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics                               Kadierdan Kaheman, J. Kutz, S. Brunton            2020-04-05 Proceedings. Mathematical, Physical, and Engineering Sciences, Proceedings of the Royal Society A 202 65 open_in_new visibility_off                  Multidimensional Approximation of Nonlinear Dynamical Systems                               Patrick Gel\u00df, Stefan Klus, J. Eisert, Christof Schutte            2018-09-07 Journal of Computational and Nonlinear Dynamics 62 77 open_in_new visibility_off                  Learning Discrepancy Models From Experimental Data                               Kadierdan Kaheman, E. Kaiser, B. Strom, J. Kutz, S. Brunton            2019-09-18 ArXiv, arXiv.org 32 65 open_in_new visibility_off                  Discovery of Physics From Data: Universal Laws and Discrepancies                               Brian M. de Silva, D. Higdon, S. Brunton, J. Kutz            2019-06-19 Frontiers in Artificial Intelligence 71 65 open_in_new visibility_off                  Data-driven discovery of partial differential equations                               S. Rudy, S. Brunton, J. Proctor, J. Kutz            2016-09-21 Science Advances 1187 65 open_in_new visibility_off                  Ensemble-SINDy: Robust sparse model discovery in the low-data, high-noise limit, with active learning and control                               Urban Fasel, J. Kutz, Bingni W. Brunton, S. Brunton            2021-11-22 Proceedings. Mathematical, Physical, and Engineering Sciences, Proceedings of the Royal Society A 161 65 open_in_new visibility_off                  Learning sparse nonlinear dynamics via mixed-integer optimization                               D. Bertsimas, Wes Gurnee            2022-06-01 Nonlinear Dynamics 29 91 open_in_new visibility_off                  A Unified Framework for Sparse Relaxed Regularized Regression: SR3                               P. Zheng, T. Askham, S. Brunton, J. Kutz, A. Aravkin            2018-07-14 IEEE Access 116 65 open_in_new Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index View recommendations"},{"location":"Symbolic%20regression/#recommended_articles","title":"Recommended articles on Symbolic regression","text":"Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index visibility_off                Discovering governing equation in structural dynamics from acceleration-only measurements                             Calvin Alvares, Souvik Chakraborty            2024-07-18 ArXiv 0 1 visibility_off                Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems                             Mohammad Amin Basiri, Sina Khanmohammadi            2024-09-02 ArXiv 0 3 visibility_off                Bayesian learning with Gaussian processes for low-dimensional representations of time-dependent nonlinear systems                             Shane A. McQuarrie, Anirban Chaudhuri, Karen Willcox, Mengwu Guo            2024-08-06 ArXiv 0 6 visibility_off                Data-driven Discovery of Delay Differential Equations with Discrete Delays                             Alessandro Pecile, N. Demo, M. Tezzele, G. Rozza, Dimitri Breda            2024-07-29 ArXiv 1 50 visibility_off                Principal Component Flow Map Learning of PDEs from Incomplete, Limited, and Noisy Data                             Victor Churchill            2024-07-15 ArXiv 0 0 visibility_off                BINDy -- Bayesian identification of nonlinear dynamics with reversible-jump Markov-chain Monte-Carlo                             M.D. Champneys, T. J. Rogers            2024-08-15 ArXiv 0 1 visibility_off                Spectrally Informed Learning of Fluid Flows                             Benjamin D. Shaffer, Jeremy R. Vorenberg, M. A. Hsieh            2024-08-26 ArXiv 0 2 visibility_off                Learning Networked Dynamical System Models with Weak Form and Graph Neural Networks                             Yin Yu, Daning Huang, Seho Park, H. Pangborn            2024-07-23 ArXiv 0 11 visibility_off                Solving Oscillator Ordinary Differential Equations via Soft-constrained Physics-informed Neural Network with Small Data                             Kai-liang Lu, Yu-meng Su, Zhuo Bi, Cheng Qiu, Wen-jun Zhang            2024-08-19 ArXiv 0 0 visibility_off                Probabilistic Decomposed Linear Dynamical Systems for Robust Discovery of Latent Neural Dynamics                             Yenho Chen, Noga Mudrik, Kyle A. Johnsen, Sankaraleengam (Sankar) Alagapan, Adam S. Charles, Christopher J. Rozell            2024-08-29 ArXiv 0 19 visibility_off                Physics-informed nonlinear vector autoregressive models for the prediction of dynamical systems                             James H. Adler, Samuel Hocking, Xiaozhe Hu, Shafiqul Islam            2024-07-25 ArXiv 0 2 visibility_off                Learning Noise-Robust Stable Koopman Operator for Control with Physics-Informed Observables                             Shahriar Akbar Sakib, Shaowu Pan            2024-08-13 ArXiv 0 0 visibility_off                Accurate data\u2010driven surrogates of dynamical systems for forward propagation of uncertainty                             Saibal De, Reese E. Jones, H. Kolla            2024-08-03 International Journal for Numerical Methods in Engineering 0 30 visibility_off                Adaptation of uncertainty-penalized Bayesian information criterion for parametric partial differential equation discovery                             Pongpisit Thanasutives, Ken-ichi Fukui            2024-08-15 ArXiv 0 3 visibility_off                Stable Sparse Operator Inference for Nonlinear Structural Dynamics                             P. D. Boef, Diana Manvelyan, Jos Maubach, W. Schilders, N. Wouw            2024-07-31 ArXiv 0 47 visibility_off                Koopman Operators in Robot Learning                             Lu Shi, Masih Haseli, Giorgos Mamakoukas, Daniel Bruder, Ian Abraham, Todd Murphey, Jorge Cortes, Konstantinos Karydis            2024-08-08 ArXiv 0 9 visibility_off                Learning Latent Space Dynamics with Model-Form Uncertainties: A Stochastic Reduced-Order Modeling Approach                             Jin Yi Yong, Rudy Geelen, Johann Guilleminot            2024-08-30 ArXiv 0 1 visibility_off                Relaxation-based schemes for on-the-fly parameter estimation in dissipative dynamical systems                             Vincent R. Martinez, Jacob Murri, J. Whitehead            2024-08-26 ArXiv 0 1 visibility_off                Learning Global Linear Representations of Truly Nonlinear Dynamics                             Thomas Breunung, F. Kogelbauer            2024-08-06 ArXiv 0 6 visibility_off                A PINN approach for the online identification and control of unknown PDEs                             Alessandro Alla, Giulia Bertaglia, Elisa Calzola            2024-08-06 ArXiv 0 1 visibility_off                Sparse identification of time delay systems via pseudospectral collocation                             Enrico Bozzo, Dimitri Breda, Muhammad Tanveer            2024-08-04 ArXiv 0 0 visibility_off                Data-driven identification of latent port-Hamiltonian systems                             J. Rettberg, Jonas Kneifl, Julius Herb, Patrick Buchfink, J. Fehr, B. Haasdonk            2024-08-15 ArXiv 0 32 visibility_off                Data-Driven Stochastic Closure Modeling via Conditional Diffusion Model and Neural Operator                             Xinghao Dong, Chuanqi Chen, Jin-Long Wu            2024-08-06 ArXiv 1 2 visibility_off                Machine Learning for the Physics of Climate                             Annalisa Bracco, Julien Brajard, Henk Dijkstra, P. Hassanzadeh, Christian Lessig, C. Monteleoni            2024-08-19 ArXiv 1 25 visibility_off                Data-driven ODE modeling of the high-frequency complex dynamics of a fluid flow                             Natsuki Tsutsumi, Kengo Nakai, Yoshitaka Saiki            2024-09-01 ArXiv 0 4 visibility_off                Extracting self-similarity from data                             Nikos Bempedelis, Luca Magri, Konstantinos Steiros            2024-07-15 ArXiv 0 0 visibility_off                Learning the Latent dynamics of Fluid flows from High-Fidelity Numerical Simulations using Parsimonious Diffusion Maps                             Alessandro Della Pia, Dimitris G. Patsatzis, Lucia Russo, C. Siettos            2024-08-05 ArXiv 0 24 visibility_off                Real-time optimal control of high-dimensional parametrized systems by deep learning-based reduced order models                             Matteo Tomasetto, Andrea Manzoni, Francesco Braghin            2024-09-09 ArXiv 0 0 visibility_off                Sampling parameters of ordinary differential equations with Langevin dynamics that satisfy constraints                             Chris Chi, J. Weare, Aaron R Dinner            2024-08-28 ArXiv 0 20 visibility_off                Modeling Latent Neural Dynamics with Gaussian Process Switching Linear Dynamical Systems                             Amber Hu, D. Zoltowski, Aditya Nair, David Anderson, Lea Duncker, Scott W. Linderman            2024-07-19 ArXiv 0 27 visibility_off                Active Symbolic Discovery of Ordinary Differential Equations via Phase Portrait Sketching                             Nan Jiang, Md. Nasim, Yexiang Xue            2024-09-02 ArXiv 0 1 visibility_off                Higher order quantum reservoir computing for non-intrusive reduced-order models                             Vinamr Jain, R. Maulik            2024-07-31 ArXiv 0 22 visibility_off                On latent dynamics learning in nonlinear reduced order modeling                             N. Farenga, S. Fresca, Simone Brivio, A. Manzoni            2024-08-27 ArXiv 0 11 visibility_off                Beyond Closure Models: Learning Chaotic-Systems via Physics-Informed Neural Operators                             Chuwei Wang, Julius Berner, Zong-Yi Li, Di Zhou, Jiayun Wang, Jane Bae, A. Anandkumar            2024-08-09 ArXiv 0 18 visibility_off                Predicting multi-parametric dynamics of externally forced oscillators using reservoir computing and minimal data                             Manish Yadav, Swati Chauhan, M. Shrimali, M. Stender            2024-08-27 ArXiv 0 19 visibility_off                Latent-EnSF: A Latent Ensemble Score Filter for High-Dimensional Data Assimilation with Sparse Observation Data                             Phillip Si, Peng Chen            2024-08-29 ArXiv 0 0 visibility_off                Stochastic Neural Simulator for Generalizing Dynamical Systems across Environments                             Liu Jiaqi, Jiaxu Cui, Jiayi Yang, Bo Yang            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 0 5 visibility_off                Non-Asymptotic Uncertainty Quantification in High-Dimensional Learning                             Frederik Hoppe, C. M. Verdun, Hannah Laus, Felix Krahmer, Holger Rauhut            2024-07-18 ArXiv 0 5 visibility_off                Self-tuning moving horizon estimation of nonlinear systems via physics-informed machine learning Koopman modeling                             Mingxue Yan, Minghao Han, A. Law, Xunyuan Yin            2024-08-07 ArXiv 0 35 visibility_off                Neural Ordinary Differential Equations for Model Order Reduction of Stiff Systems                             Matteo Caldana, J. Hesthaven            2024-08-12 ArXiv 0 64 visibility_off                Optimal Experimental Design for Universal Differential Equations                             Christoph Plate, Carl Julius Martensen, Sebastian Sager            2024-08-13 ArXiv 0 1 visibility_off                State Space Kriging model for emulating complex nonlinear dynamical systems under stochastic excitation                             Kai Chenga, Iason Papaioannoua, MengZe Lyub, Daniel Straub            2024-09-04 ArXiv 0 0 visibility_off                A Physics-Informed Machine Learning Approach for Solving Distributed Order Fractional Differential Equations                             A. Aghaei            2024-09-05 ArXiv 0 5 visibility_off                A Regularized Physics-Informed Neural Network to Support Data-Driven Nonlinear Constrained Optimization                             Diego Armando Perez-Rosero, A. \u00c1lvarez-Meza, G. Castellanos-Dom\u00ednguez            2024-07-18 Comput. 0 25 visibility_off                Physics-informed Discovery of State Variables in Second-Order and Hamiltonian Systems                             F\u00e9lix Chavelli, Zi-Yu Khoo, Dawen Wu, Jonathan Sze Choong Low, St\u00e9phane Bressan            2024-08-21 ArXiv 0 3 visibility_off                Practical Guidelines for Data-driven Identification of Lifted Linear Predictors for Control                             Loi Do, Adam Uchytil, Zdenvek Hur'ak            2024-08-02 ArXiv 1 2 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index"},{"location":"Time-series%20forecasting/","title":"Time-series forecasting","text":"<p> This page was last updated on 2024-09-11 16:11:43 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p> </p> <p> </p>"},{"location":"Time-series%20forecasting/#manually_curated_articles","title":"Manually curated articles on Time-series forecasting","text":"Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index                View recommendations                visibility_off                  A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection                               Ming Jin, Huan Yee Koh, Qingsong Wen, Daniele Zambon, C. Alippi, G. I. Webb, Irwin King, Shirui Pan            2023-07-07 IEEE transactions on pattern analysis and machine intelligence 64 50 open_in_new visibility_off                  Graph-Guided Network for Irregularly Sampled Multivariate Time Series                               Xiang Zhang, M. Zeman, Theodoros Tsiligkaridis, M. Zitnik            2021-10-11 ArXiv, International Conference on Learning Representations 73 47 open_in_new visibility_off                  Taming Local Effects in Graph-based Spatiotemporal Forecasting                               Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-02-08 ArXiv, Neural Information Processing Systems 18 50 open_in_new visibility_off                  Sparse Graph Learning from Spatiotemporal Time Series                               Andrea Cini, Daniele Zambon, C. Alippi            2022-05-26 J. Mach. Learn. Res., Journal of machine learning research 11 50 open_in_new visibility_off                  Graph Deep Learning for Time Series Forecasting                               Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-10-24 ArXiv, arXiv.org 6 50 open_in_new visibility_off                  Large Language Models Are Zero-Shot Time Series Forecasters                               Nate Gruver, Marc Finzi, Shikai Qiu, Andrew Gordon Wilson            2023-10-11 ArXiv, Neural Information Processing Systems 129 14 open_in_new visibility_off                  Graph-Mamba: Towards Long-Range Graph Sequence Modeling with Selective State Spaces                               Chloe X. Wang, Oleksii Tsepa, Jun Ma, Bo Wang            2024-02-01 ArXiv, arXiv.org 50 5 open_in_new visibility_off                  A decoder-only foundation model for time-series forecasting                               Abhimanyu Das, Weihao Kong, Rajat Sen, Yichen Zhou            2023-10-14 ArXiv, International Conference on Machine Learning 51 14 open_in_new visibility_off                  Unified Training of Universal Time Series Forecasting Transformers                               Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, Doyen Sahoo            2024-02-04 ArXiv, International Conference on Machine Learning 32 22 open_in_new visibility_off                  Time-LLM: Time Series Forecasting by Reprogramming Large Language Models                               Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen            2023-10-03 ArXiv, International Conference on Learning Representations 134 9 open_in_new visibility_off                  Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series                               Vijay Ekambaram, Arindam Jati, Nam H. Nguyen, Pankaj Dayama, Chandra Reddy, Wesley M. Gifford, Jayant Kalagnanam            2024-01-08 ArXiv, arXiv.org 2 2 open_in_new visibility_off                  Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistency                               Xiang Zhang, Ziyuan Zhao, Theodoros Tsiligkaridis, M. Zitnik            2022-06-17 ArXiv, Neural Information Processing Systems 176 47 open_in_new visibility_off                  Domain Adaptation for Time Series Under Feature and Label Shifts                               Huan He, Owen Queen, Teddy Koker, Consuelo Cuevas, Theodoros Tsiligkaridis, M. Zitnik            2023-02-06 ArXiv, DBLP 26 47 open_in_new visibility_off                  AZ-whiteness test: a test for signal uncorrelation on spatio-temporal graphs                               Daniele Zambon, C. Alippi            None DBLP 6 50 open_in_new visibility_off                  Graph state-space models                               Daniele Zambon, Andrea Cini, L. Livi, C. Alippi            2023-01-04 ArXiv, arXiv.org 3 50 open_in_new visibility_off                  UNITS: A Unified Multi-Task Time Series Model                               Shanghua Gao, Teddy Koker, Owen Queen, Thomas Hartvigsen, Theodoros Tsiligkaridis, M. Zitnik            2024-02-29 ArXiv 2 47 open_in_new Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index View recommendations"},{"location":"Time-series%20forecasting/#recommended_articles","title":"Recommended articles on Time-series forecasting","text":"Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index visibility_off                LeRet: Language-Empowered Retentive Network for Time Series Forecasting                             Qihe Huang, Zhen-Qiang Zhou, Kuo Yang, Gengyu Lin, Zhongchao Yi, Yang Wang            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 0 3 visibility_off                TimeDiT: General-purpose Diffusion Transformers for Time Series Foundation Model                             Defu Cao, Wen Ye, Yizhou Zhang, Yan Liu            2024-09-03 ArXiv 0 9 visibility_off                PRformer: Pyramidal Recurrent Transformer for Multivariate Time Series Forecasting                             Yongbo Yu, Weizhong Yu, Feiping Nie, Xuelong Li            2024-08-20 ArXiv 0 10 visibility_off                A federated large language model for long-term time series forecasting                             Raed Abdel Sater, A. B. Hamza            2024-07-30 ArXiv 0 1 visibility_off                Deep Time Series Models: A Comprehensive Survey and Benchmark                             Yuxuan Wang, Haixu Wu, Jiaxiang Dong, Yong Liu, Mingsheng Long, Jianmin Wang            2024-07-18 ArXiv 3 66 visibility_off                Simplified Mamba with Disentangled Dependency Encoding for Long-Term Time Series Forecasting                             Zixuan Weng, Jindong Han, Wenzhao Jiang, Hao Liu            2024-08-22 ArXiv 0 10 visibility_off                ISMRNN: An Implicitly Segmented RNN Method with Mamba for Long-Term Time Series Forecasting                             Gaoxiang Zhao, Li Zhou, Xiaoqiang Wang            2024-07-15 ArXiv 0 0 visibility_off                Skip-Timeformer: Skip-Time Interaction Transformer for Long Sequence Time-Series Forecasting                             Wenchang Zhang, Hua Wang, Fan Zhang            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 0 4 visibility_off                PLUTUS: A Well Pre-trained Large Unified Transformer can Unveil Financial Time Series Regularities                             Yuanjian Xu, Anxian Liu, Jianing Hao, Zhenzhuo Li, Shichang Meng, Guang Zhang            2024-08-19 ArXiv 0 0 visibility_off                Agentic Retrieval-Augmented Generation for Time Series Analysis                             Chidaksh Ravuru, Sagar Srinivas Sakhinana, Venkataramana Runkana            2024-08-18 ArXiv 0 3 visibility_off                Decoupled Invariant Attention Network for Multivariate Time-series Forecasting                             Haihua Xu, Wei Fan, Kun Yi, Pengyang Wang            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 0 4 visibility_off                Scalable Transformer for High Dimensional Multivariate Time Series Forecasting                             Xin Zhou, Weiqing Wang, W. Buntine, Shilin Qu, Abishek Sriramulu, Weicong Tan, Christoph Bergmeir            2024-08-08 ArXiv 0 16 visibility_off                AutoCTS++: zero-shot joint neural architecture and hyperparameter search for correlated time series forecasting                             Xinle Wu, Xingjian Wu, Bin Yang, Lekui Zhou, Chenjuan Guo, Xiangfei Qiu, Jilin Hu, Zhenli Sheng, Christian S. Jensen            2024-07-30 VLDB J. 0 27 visibility_off                Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting                             Alireza Jafari, Geoffrey C. Fox, John B. Rundle, A. Donnellan, L. G. Ludwig            2024-08-21 ArXiv 0 31 visibility_off                xLSTMTime : Long-term Time Series Forecasting With xLSTM                             Musleh Alharthi, A. Mahmood            2024-07-14 ArXiv 0 22 visibility_off                sTransformer: A Modular Approach for Extracting Inter-Sequential and Temporal Information for Time-Series Forecasting                             Jiaheng Yin, Zhengxin Shi, Jianshen Zhang, Xiaoming Lin, Yulin Huang, Yongzhi Qi, Wei Qi            2024-08-19 ArXiv 0 2 visibility_off                Unleash The Power of Pre-Trained Language Models for Irregularly Sampled Time Series                             Weijiao Zhang, Chenlong Yin, Hao Liu, Hui Xiong            2024-08-12 ArXiv 0 6 visibility_off                DRFormer: Multi-Scale Transformer Utilizing Diverse Receptive Fields for Long Time-Series Forecasting                             Ruixin Ding, Yuqi Chen, Yu-Ting Lan, Wei Zhang            2024-08-05 ArXiv 0 0 visibility_off                DAM: Towards A Foundation Model for Time Series Forecasting                             Luke Darlow, Qiwen Deng, Ahmed Hassan, Martin Asenov, Rajkarn Singh, Artjom Joosen, Adam Barker, A. Storkey            2024-07-25 ArXiv 0 43 visibility_off                Mamba or Transformer for Time Series Forecasting? Mixture of Universals (MoU) Is All You Need                             Sijia Peng, Yun Xiong, Yangyong Zhu, Zhiqiang Shen            2024-08-28 ArXiv 0 19 visibility_off                FMamba: Mamba based on Fast-attention for Multivariate Time-series Forecasting                             Shusen Ma, Yu Kang, Peng Bai, Yunan Zhao            2024-07-20 ArXiv 0 1 visibility_off                DTFormer: A Transformer-Based Method for Discrete-Time Dynamic Graph Representation Learning                             Xi Chen, Yun Xiong, Siwei Zhang, Jiawei Zhang, Yao Zhang, Shiyang Zhou, Xixi Wu, Mingyang Zhang, Tengfei Liu, Weiqiang Wang            2024-07-26 ArXiv 0 4 visibility_off                Unlocking the Power of LSTM for Long Term Time Series Forecasting                             Yaxuan Kong, Zepu Wang, Yuqi Nie, Tian Zhou, Stefan Zohren, Yuxuan Liang, Peng Sun, Qingsong Wen            2024-08-19 ArXiv 0 5 visibility_off                Multiscale Representation Enhanced Temporal Flow Fusion Model for Long-Term Workload Forecasting                             Shiyu Wang, Zhixuan Chu, Yinbo Sun, Yu Liu, Yuliang Guo, Yang Chen, Huiyang Jian, Lintao Ma, Xingyu Lu, Jun Zhou            2024-07-29 ArXiv 0 4 visibility_off                An NCDE-based Framework for Universal Representation Learning of Time Series                             Zihang Liu, Bowen Du, Junchen Ye, Xianqing Wen, Leilei Sun            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 0 26 visibility_off                Omni-Dimensional Frequency Learner for General Time Series Analysis                             Xianing Chen.Hanting Chen, Hanting Chen, Hailin Hu            2024-07-15 ArXiv 0 1 visibility_off                An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting                             Rui Cao, Qiao Wang            2024-08-09 ArXiv 0 3 visibility_off                MedTsLLM: Leveraging LLMs for Multimodal Medical Time Series Analysis                             Nimeesha Chan, Felix Parker, William Bennett, Tianyi Wu, Mung Yao Jia, James Fackler, Kimia Ghobadi            2024-08-14 ArXiv 0 1 visibility_off                Long Input Sequence Network for Long Time Series Forecasting                             Chao Ma, Yikai Hou, Xiang Li, Yinggang Sun, Haining Yu            2024-07-18 ArXiv 0 1 visibility_off                SDformer: Transformer with Spectral Filter and Dynamic Attention for Multivariate Time Series Long-term Forecasting                             Ziyu Zhou, Gengyu Lyu, Yiming Huang, Zihao Wang, Ziyu Jia, Zhen Yang            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 0 11 visibility_off                Revisiting Attention for Multivariate Time Series Forecasting                             Haixiang Wu            2024-07-18 ArXiv 0 0 visibility_off                Transformer-Based Spatiotemporal Graph Diffusion Convolution Network for Traffic Flow Forecasting                             Siwei Wei, Yang Yang, Donghua Liu, Ke Deng, Chunzhi Wang            2024-08-09 Electronics 0 2 visibility_off                TempoKGAT: A Novel Graph Attention Network Approach for Temporal Graph Analysis                             L. Sasal, Daniel Busby, Abdenour Hadid            2024-08-29 ArXiv 0 2 visibility_off                Heterogeneous Graph Transformer with Poly-Tokenization                             Zhiyuan Lu, Yuan Fang, Cheng Yang, Chuan Shi            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 1 4 visibility_off                Robust Multivariate Time Series Forecasting against Intra- and Inter-Series Transitional Shift                             Hui He, Qi Zhang, Kun Yi, Xiaojun Xue, Shoujin Wang, Liang Hu, Longbin Cao            2024-07-18 ArXiv 0 6 visibility_off                DLFormer: Enhancing Explainability in Multivariate Time Series Forecasting using Distributed Lag Embedding                             Younghwi Kim, Dohee Kim, Sunghyun Sim            2024-08-29 ArXiv 0 4 visibility_off                Multi-View Neural Differential Equations for Continuous-Time Stream Data in Long-Term Traffic Forecasting                             Zibo Liu, Zhe Jiang, Shigang Chen            2024-08-12 ArXiv 0 2 visibility_off                Contrastive Representation Learning for Dynamic Link Prediction in Temporal Networks                             Amirhossein Nouranizadeh, Fatemeh Tabatabaei Far, Mohammad Rahmati            2024-08-22 ArXiv 0 2 visibility_off                DSTLNet: Dynamic Spatial-Temporal Correlation Learning Network for Traffic Sensor Signal Prediction                             Yuxiang Shan, Hailiang Lu, Weidong Lou            2024-08-08 Sensors and Materials 0 3 visibility_off                Survey and Taxonomy: The Role of Data-Centric AI in Transformer-Based Time Series Forecasting                             Jingjing Xu, Caesar Wu, Yuan-Fang Li, Gr\u00e9goire Danoy, Pascal Bouvry            2024-07-29 ArXiv 0 23 visibility_off                Linear Attention is Enough in Spatial-Temporal Forecasting                             Xinyu Ning            2024-08-17 ArXiv 0 0 visibility_off                SpecAR-Net: Spectrogram Analysis and Representation Network for Time Series                             Yi Dong, Liwen Zhang, Youcheng Zhang, Shi Peng, Wen Chen, Zhe Ma            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 0 0 visibility_off                DyGMamba: Efficiently Modeling Long-Term Temporal Dependency on Continuous-Time Dynamic Graphs with State Space Models                             Zifeng Ding, Yifeng Li, Yuan He, Antonio Norelli, Jingcheng Wu, Volker Tresp, Yunpu Ma, Michael Bronstein            2024-08-08 ArXiv 0 22 visibility_off                Modeling epidemic dynamics using Graph Attention based Spatial Temporal networks                             Xiaofeng Zhu, Yi Zhang, Haoru Ying, Huanning Chi, Guanqun Sun, Lingxia Zeng            2024-07-15 PLOS ONE 0 0 visibility_off                An Evaluation of Deep Learning Models for Stock Market Trend Prediction                             Gonzalo Lopez Gil, Paul Duhamel-Sebline, Andrew McCarren            2024-08-22 ArXiv 0 0 visibility_off                SCAT: A Time Series Forecasting with Spectral Central Alternating Transformers                             Chengjie Zhou, Chao Che, Pengfei Wang, Qiang Zhang            2024-08-01 Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence 0 8 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index"},{"location":"recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378/","title":"06a0ba437d41a7c82c08a9636a4438c1b5031378","text":""},{"location":"recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378/#_1","title":"06a0ba437d41a7c82c08a9636a4438c1b5031378","text":"<p> This page was last updated on 2024-09-11 16:11:58 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Inferring Biological Networks by Sparse Identification of Nonlinear Dynamics Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Distilling identifiable and interpretable dynamic models from biological data              Gemma Massonis, A. F. Villaverde, J. Banga            2023-03-15 PLOS Computational Biology 3 55 visibility_off Characterization of Biologically Relevant Network Structures form Time-series Data              Z. Tuza, G. Stan            2018-09-24 2018 IEEE Conference on Decision and Control (CDC) 3 35 visibility_off Reconstruction of arbitrary biochemical reaction networks: A compressive sensing approach              W. Pan, Ye Yuan, G. Stan            2012-05-08 2012 IEEE 51st IEEE Conference on Decision and Control (CDC) 33 35 visibility_off Inferring sparse networks for noisy transient processes              H. M. Tran, S. Bukkapatnam            2016-02-26 Scientific Reports 6 34 visibility_off Koopman Operators for Generalized Persistence of Excitation Conditions for Nonlinear Systems              N. Boddupalli, A. Hasnain, S. Nandanoori, Enoch Yeung            2019-06-25 2019 IEEE 58th Conference on Decision and Control (CDC) 5 17 visibility_off Learning sparse nonlinear dynamics via mixed-integer optimization              D. Bertsimas, Wes Gurnee            2022-06-01 Nonlinear Dynamics 29 91 visibility_off Inverse problems in systems biology              H. Engl, Christoph Flamm, P. K\u00fcgler, James Lu, S. M\u00fcller, P. Schuster            2009-12-01 Inverse Problems 136 48 visibility_off Set-base dynamical parameter estimation and model invalidation for biochemical reaction networks              Philipp Rumschinski, S. Borchers, S. Bosio, R. Weismantel, R. Findeisen            2010-05-25 BMC Systems Biology 67 47 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/0acd117521ef5aafb09fed02ab415523b330b058/","title":"0acd117521ef5aafb09fed02ab415523b330b058","text":""},{"location":"recommendations/0acd117521ef5aafb09fed02ab415523b330b058/#_1","title":"0acd117521ef5aafb09fed02ab415523b330b058","text":"<p> This page was last updated on 2024-09-11 16:12:06 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Data-driven discovery of partial differential equations Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Machine discovery of partial differential equations from spatiotemporal data: A sparse Bayesian learning framework.              Ye Yuan, Xiuting Li, Liang Li, Frank J. Jiang, Xiuchuan Tang, Fumin Zhang, Jorge Gon\u00e7alves, Henning U. Voss, Han Ding, J\u00fcrgen Kurths            2023-11-01 Chaos 3 11 visibility_off Machine Discovery of Partial Differential Equations from Spatiotemporal Data              Ye Yuan, Junlin Li, Liang Li, Frank Jiang, Xiuchuan Tang, Fumin Zhang, Sheng Liu, J. Gon\u00e7alves, H. Voss, Xiuting Li, J. Kurths, Han Ding            2019-09-15 ArXiv 9 107 visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3256 65 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 8 2 visibility_off Physics-informed learning of governing equations from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 Nature Communications 250 12 visibility_off Sparsistent Model Discovery              Georges Tod, G. Both, R. Kusters            2021-06-22 ArXiv 1 12 visibility_off Supplementary material from \"Learning partial differential equations via data discovery and sparse optimization\"              Hayden Schaeffer            2017-01-16 0 15 visibility_off Data-Driven discovery of governing physical laws and their parametric dependencies in engineering, physics and biology              J. Kutz, Samuel H. Rudy, A. Alla, S. Brunton            2017-12-01 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP) 12 65 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b/","title":"0d01d21137a5af9f04e4b16a55a0f732cb8a540b","text":""},{"location":"recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b/#_1","title":"0d01d21137a5af9f04e4b16a55a0f732cb8a540b","text":"<p> This page was last updated on 2024-09-11 16:11:15 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Sparse Graph Learning from Spatiotemporal Time Series Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Graph Deep Learning for Time Series Forecasting              Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-10-24 ArXiv 6 50 visibility_off Sparsity exploitation via discovering graphical models in multi-variate time-series forecasting              Ngoc-Dung Do, T. Hy, D. Nguyen            2023-06-29 ArXiv 0 7 visibility_off TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting              Nancy R. Xu, Chrysoula Kosma, M. Vazirgiannis            2023-07-27 ArXiv 0 54 visibility_off Balanced Graph Structure Learning for Multivariate Time Series Forecasting              Weijun Chen, Yanze Wang, Chengshuo Du, Zhenglong Jia, Feng Liu, Ran Chen            2022-01-24 ArXiv 0 10 visibility_off A Study of Joint Graph Inference and Forecasting              Daniel Z\u00fcgner, Fran\u00e7ois-Xavier Aubet, Victor Garcia Satorras, Tim Januschowski, Stephan G\u00fcnnemann, Jan Gasthaus            2021-09-10 ArXiv 12 46 visibility_off Regularized Graph Structure Learning with Semantic Knowledge for Multi-variates Time-Series Forecasting              Hongyuan Yu, Ting Li, Weichen Yu, Jianguo Li, Yan Huang, Liang Wang, A. Liu            2022-07-01 ArXiv, DBLP 40 34 visibility_off Discrete Graph Structure Learning for Forecasting Multiple Time Series              Chao Shang, Jie Chen, J. Bi            2021-01-18 ArXiv 179 37 visibility_off Learning Graph Structures and Uncertainty for Accurate and Calibrated Time-series Forecasting              Harshavardhan Kamarthi, Lingkai Kong, Alexander Rodr\u00edguez, Chao Zhang, B. A. Prakash            2024-07-02 ArXiv 0 9 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/123acfbccca0460171b6b06a4012dbb991cde55b/","title":"123acfbccca0460171b6b06a4012dbb991cde55b","text":""},{"location":"recommendations/123acfbccca0460171b6b06a4012dbb991cde55b/#_1","title":"123acfbccca0460171b6b06a4012dbb991cde55b","text":"<p> This page was last updated on 2024-09-11 16:11:17 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Large Language Models Are Zero-Shot Time Series Forecasters Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Chronos: Learning the Language of Time Series              Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang            2024-03-12 ArXiv 28 18 visibility_off An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting              Rui Cao, Qiao Wang            2024-08-09 ArXiv 0 3 visibility_off AutoTimes: Autoregressive Time Series Forecasters via Large Language Models              Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-02-04 ArXiv 4 66 visibility_off Are Language Models Actually Useful for Time Series Forecasting?              Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen            2024-06-22 ArXiv 1 1 visibility_off Timer: Generative Pre-trained Transformers Are Large Time Series Models              Yong Liu, Haoran Zhang, Chenyu Li, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-02-04 ArXiv, DBLP 4 66 visibility_off TimeGPT-1              Azul Garza, Cristian Challu, Max Mergenthaler-Canseco            2023-10-05 ArXiv 43 1 visibility_off MultiCast: Zero-Shot Multivariate Time Series Forecasting Using LLMs              Georgios Chatzigeorgakidis, Konstantinos Lentzos, Dimitrios Skoutas            2024-05-13 2024 IEEE 40th International Conference on Data Engineering Workshops (ICDEW) 0 7 visibility_off Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilities              Mingyu Jin, Hua Tang, Chong Zhang, Qinkai Yu, Chengzhi Liu, Suiyuan Zhu, Yongfeng Zhang, Mengnan Du            2024-02-16 ArXiv 3 6 visibility_off Time-LLM: Time Series Forecasting by Reprogramming Large Language Models              Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen            2023-10-03 ArXiv 134 9 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277/","title":"16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277","text":""},{"location":"recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277/#_1","title":"16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277","text":"<p> This page was last updated on 2024-09-11 16:11:30 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Time-LLM: Time Series Forecasting by Reprogramming Large Language Models Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Time-FFM: Towards LM-Empowered Federated Foundation Model for Time Series Forecasting              Qingxiang Liu, Xu Liu, Chenghao Liu, Qingsong Wen, Yuxuan Liang            2024-05-23 ArXiv 0 6 visibility_off Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilities              Mingyu Jin, Hua Tang, Chong Zhang, Qinkai Yu, Chengzhi Liu, Suiyuan Zhu, Yongfeng Zhang, Mengnan Du            2024-02-16 ArXiv 3 6 visibility_off LSTPrompt: Large Language Models as Zero-Shot Time Series Forecasters by Long-Short-Term Prompting              Haoxin Liu, Zhiyuan Zhao, Jindong Wang, Harshavardhan Kamarthi, B. A. Prakash            2024-02-25 ArXiv 6 7 visibility_off PromptCast: A New Prompt-Based Learning Paradigm for Time Series Forecasting              Hao Xue, Flora D.Salim            2022-09-20 IEEE Transactions on Knowledge and Data Engineering 63 17 visibility_off Understanding Different Design Choices in Training Large Time Series Models              Yu-Neng Chuang, Songchen Li, Jiayi Yuan, Guanchu Wang, Kwei-Herng Lai, Leisheng Yu, Sirui Ding, Chia-yuan Chang, Qiaoyu Tan, D. Zha, Xia Hu            2024-06-20 ArXiv 1 23 visibility_off A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model              Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung            2024-05-03 ArXiv 4 47 visibility_off In-context Time Series Predictor              Jiecheng Lu, Yan Sun, Shihao Yang            2024-05-23 ArXiv 0 1 visibility_off LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters              Ching Chang, Wenjie Peng, Tien-Fu Chen            2023-08-16 ArXiv 16 2 visibility_off Are Language Models Actually Useful for Time Series Forecasting?              Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen            2024-06-22 ArXiv 1 1 visibility_off A decoder-only foundation model for time-series forecasting              Abhimanyu Das, Weihao Kong, Rajat Sen, Yichen Zhou            2023-10-14 ArXiv 51 14 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f/","title":"1df04f33a8ef313cc2067147dbb79c3ca7c5c99f","text":""},{"location":"recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f/#_1","title":"1df04f33a8ef313cc2067147dbb79c3ca7c5c99f","text":"<p> This page was last updated on 2024-09-11 16:11:17 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Graph-Mamba: Towards Long-Range Graph Sequence Modeling with Selective State Spaces Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off HeteGraph-Mamba: Heterogeneous Graph Learning via Selective State Space Model              Zhenyu Pan, Yoonsung Jeong, Xiaoda Liu, Han Liu            2024-05-22 ArXiv 0 2 visibility_off Graph Mamba: Towards Learning on Graphs with State Space Models              Ali Behrouz, Farnoosh Hashemi            2024-02-13 ArXiv, DBLP 33 9 visibility_off Learning Long Range Dependencies on Graphs via Random Walks              Dexiong Chen, Till Hendrik Schulz, Karsten Borgwardt            2024-06-05 ArXiv 0 8 visibility_off What Can We Learn from State Space Models for Machine Learning on Graphs?              Yinan Huang, Siqi Miao, Pan Li            2024-06-09 ArXiv 1 3 visibility_off A Scalable and Effective Alternative to Graph Transformers              Kaan Sancak, Zhigang Hua, Jin Fang, Yan Xie, Andrey Malevich, Bo Long, M. F. Balin, \u00dcmit V. \u00c7ataly\u00fcrek            2024-06-17 ArXiv 0 7 visibility_off Hierarchical Graph Transformer with Adaptive Node Sampling              Zaixin Zhang, Qi Liu, Qingyong Hu, Cheekong Lee            2022-10-08 ArXiv 59 19 visibility_off Deformable Graph Transformer              Jinyoung Park, Seongjun Yun, Hyeon-ju Park, Jaewoo Kang, Jisu Jeong, KyungHyun Kim, Jung-Woo Ha, Hyunwoo J. Kim            2022-06-29 ArXiv 6 28 visibility_off LazyGNN: Large-Scale Graph Neural Networks via Lazy Propagation              Rui Xue, Haoyu Han, MohamadAli Torkamani, Jian Pei, Xiaorui Liu            2023-02-03 ArXiv 9 7 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68/","title":"279cd637b7e38bba1dd8915b5ce68cbcacecbe68","text":""},{"location":"recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68/#_1","title":"279cd637b7e38bba1dd8915b5ce68cbcacecbe68","text":"<p> This page was last updated on 2024-09-11 16:11:34 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Graph state-space models Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Probabilistic Recurrent State-Space Models              Andreas Doerr, Christian Daniel, Martin Schiegg, D. Nguyen-Tuong, S. Schaal, Marc Toussaint, Sebastian Trimpe            2018-01-31 ArXiv, DBLP, MAG 111 93 visibility_off Interpretable Latent Variables in Deep State Space Models              Haoxuan Wu, David S. Matteson, M. Wells            2022-03-03 ArXiv 0 40 visibility_off Stanza: A Nonlinear State Space Model for Probabilistic Inference in Non-Stationary Time Series              Anna K. Yanchenko, S. Mukherjee            2020-06-11 ArXiv 6 31 visibility_off Sparse Graph Learning from Spatiotemporal Time Series              Andrea Cini, Daniele Zambon, C. Alippi            2022-05-26 J. Mach. Learn. Res. 11 50 visibility_off Multi-Knowledge Fusion Network for Time Series Representation Learning              Sagar Srinivas Sakhinana, Shivam Gupta, Krishna Sai Sudhir Aripirala, Venkataramana Runkana            2024-08-22 ArXiv 0 3 visibility_off State Space Models on Temporal Graphs: A First-Principles Study              Jintang Li, Ruofan Wu, Xinzhou Jin, Boqun Ma, Liang Chen, Zibin Zheng            2024-06-03 ArXiv 1 11 visibility_off Effectively Modeling Time Series with Simple Discrete State Spaces              Michael Zhang, Khaled Kamal Saab, Michael Poli, Tri Dao, Karan Goel, Christopher R\u00e9            2023-03-16 ArXiv 30 44 visibility_off Structured Inference Networks for Nonlinear State Space Models              R. G. Krishnan, Uri Shalit, D. Sontag            2016-09-30 ArXiv, DBLP, MAG 429 48 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196/","title":"2b2aa13d4959073f61ad70555bc8c7da7d116196","text":""},{"location":"recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196/#_1","title":"2b2aa13d4959073f61ad70555bc8c7da7d116196","text":"<p> This page was last updated on 2024-09-11 16:12:04 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Multidimensional Approximation of Nonlinear Dynamical Systems Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off A dynamic mode decomposition extension for the forecasting of parametric dynamical systems              Francesco Andreuzzi, N. Demo, G. Rozza            2021-10-18 ArXiv 18 50 visibility_off Dynamic tensor time series modeling and analysis              A. Surana, G. Patterson, I. Rajapakse            2016-12-01 2016 IEEE 55th Conference on Decision and Control (CDC) 8 24 visibility_off Tensor Train Based Higher Order\u00a0Dynamic Mode Decomposition for Dynamical Systems              Keren Li, S. Utyuzhnikov            2023-04-11 SSRN Electronic Journal 3 12 visibility_off Modeling of dynamical systems through deep learning              P. Rajendra, V. Brahmajirao            2020-11-22 Biophysical Reviews 35 5 visibility_off A dynamical systems based framework for dimension reduction              Ryeongkyung Yoon, B. Osting            2022-04-18 ArXiv 1 19 visibility_off Learning Data-Driven Model of Damped Coupled Oscillators from System Impulse Response              Jacob Fabro, G. Vogl, Yongzhi Qu, Reese Eischens            2022-10-28 Annual Conference of the PHM Society 0 19 visibility_off Deep Neural Networks for Nonlinear Model Order Reduction of Unsteady Flows              Hamidreza Eivazi, H. Veisi, M. H. Naderi, V. Esfahanian            2020-07-02 ArXiv 129 23 visibility_off Nonlinear system identification with regularized Tensor Network B-splines              Ridvan Karagoz, Kim Batselier            2020-03-17 Autom. 18 4 visibility_off Data-driven prediction in dynamical systems: recent developments              Amin Ghadami, B. Epureanu            2022-06-20 Philosophical transactions. Series A, Mathematical, physical, and engineering sciences 43 33 visibility_off An Online Manifold Learning Approach for Model Reduction of Dynamical Systems              Liqian Peng, K. Mohseni            2012-10-10 SIAM J. Numer. Anal. 22 40 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220/","title":"35e2571c17246577e0bc1b9de57a314c3b60e220","text":""},{"location":"recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220/#_1","title":"35e2571c17246577e0bc1b9de57a314c3b60e220","text":"<p> This page was last updated on 2024-09-11 16:12:05 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Discovery of Physics From Data: Universal Laws and Discrepancies Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off A Unified Sparse Optimization Framework to Learn Parsimonious Physics-Informed Models From Data              Kathleen P. Champion, P. Zheng, A. Aravkin, S. Brunton, J. Kutz            2019-06-25 IEEE Access 98 65 visibility_off Machine-Learning Non-Conservative Dynamics for New-Physics Detection              Ziming Liu, Bohan Wang, Qi Meng, Wei Chen, M. Tegmark, Tie-Yan Liu            2021-05-31 Physical review. E 14 82 visibility_off Discovering conservation laws from trajectories via machine learning              Seungwoong Ha, Hawoong Jeong            2021-02-08 ArXiv 10 43 visibility_off The Lie Detector.              A. Young, A. Lawrie            2019-12-13 arXiv: Signal Processing 0 18 visibility_off Physics-Informed Machine Learning for Modeling and Control of Dynamical Systems              Truong X. Nghiem, J\u00e1n Drgo\u0148a, Colin N. Jones, Zolt\u00e1n Nagy, Roland Schwan, Biswadip Dey, A. Chakrabarty, S. D. Cairano, J. Paulson, Andrea Carron, M. Zeilinger, Wenceslao Shaw-Cortez, D. Vrabie            2023-05-31 2023 American Control Conference (ACC) 20 38 visibility_off Scale-invariant Learning by Physics Inversion              Philipp Holl, V. Koltun, Nils Thuerey            2021-09-30 ArXiv, DBLP 5 105 visibility_off Learning continuous models for continuous physics              A. Krishnapriyan, A. Queiruga, N. Benjamin Erichson, Michael W. Mahoney            2022-02-17 Communications Physics 23 30 visibility_off Machine Learning Conservation Laws from Trajectories.              Ziming Liu, Max Tegmark            2021-05-06 Physical review letters 87 82 visibility_off From data to conservation laws              F. Chirigati            2023-09-01 Nature Computational Science 0 17 visibility_off Physics-informed regularization and structure preservation for learning stable reduced models from data with operator inference              N. Sawant, B. Kramer, B. Peherstorfer            2021-07-06 ArXiv 23 27 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/3c9961153493370500020c81527b3548c96f81e0/","title":"3c9961153493370500020c81527b3548c96f81e0","text":""},{"location":"recommendations/3c9961153493370500020c81527b3548c96f81e0/#_1","title":"3c9961153493370500020c81527b3548c96f81e0","text":"<p> This page was last updated on 2024-09-11 16:11:55 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Data-driven discovery of coordinates and governing equations Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3256 65 visibility_off Physics-informed learning of governing equations from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 Nature Communications 250 12 visibility_off Deep learning of physical laws from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 ArXiv 19 12 visibility_off Modeling of dynamical systems through deep learning              P. Rajendra, V. Brahmajirao            2020-11-22 Biophysical Reviews 35 5 visibility_off Bayesian autoencoders for data-driven discovery of coordinates, governing equations and fundamental constants              Liyao (Mars) Gao, J. Kutz            2022-11-19 Proceedings of the Royal Society A 12 31 visibility_off Symbolic regression via neural networks.              N. Boddupalli, T. Matchen, J. Moehlis            2023-08-01 Chaos 2 37 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 8 2 visibility_off Discovering sparse interpretable dynamics from partial observations              Peter Y. Lu, Joan Ari\u00f1o Bernad, M. Solja\u010di\u0107            2021-07-22 Communications Physics 17 94 visibility_off Sparse Estimation for Hamiltonian Mechanics              Yuya Note, Masahito Watanabe, Hiroaki Yoshimura, Takaharu Yaguchi, Toshiaki Omori            2024-03-25 Mathematics 0 11 visibility_off Uncovering Closed-form Governing Equations of Nonlinear Dynamics from Videos              Lele Luan, Yang Liu, Hao Sun            2021-06-09 ArXiv 0 12 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2/","title":"3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2","text":""},{"location":"recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2/#_1","title":"3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2","text":"<p> This page was last updated on 2024-09-11 16:11:56 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Chaos as an intermittently forced linear system Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Featurizing Koopman mode decomposition for robust forecasting              D. Aristoff, J. Copperman, Nathan Mankovich, Alexander Davies            2023-12-14 The Journal of Chemical Physics 0 12 visibility_off Deep learning delay coordinate dynamics for chaotic attractors from partial observable data              Charles D. Young, M. Graham            2022-11-20 Physical review. E 9 51 visibility_off Discovering governing equations from partial measurements with deep delay autoencoders              Joseph Bakarji, Kathleen P. Champion, J. Nathan Kutz, S. Brunton            2022-01-13 Proceedings of the Royal Society A 70 65 visibility_off DEFM: Delay-embedding-based forecast machine for time series forecasting by spatiotemporal information transformation.              Hao Peng, Pei Chen, R. Liu            2020-05-16 Chaos 1 80 visibility_off DEFM: Delay-embedding-based forecast machine for time series forecasting by spatiotemporal information transformation.              Hao Peng, Pei Chen, R. Liu            2020-05-16 Chaos 1 80 visibility_off Detecting chaos in lineage-trees: A deep learning approach              H. Rappeport, Irit Levin Reisman, Naftali Tishby, N. Balaban            2021-06-08 ArXiv 3 56 visibility_off Cluster-based network modeling\u2014From snapshots to complex dynamical systems              Daniel Fernex, B. R. Noack, R. Semaan            2021-06-01 Science Advances 44 49 visibility_off Cluster-based network modeling -- automated robust modeling of complex dynamical systems              Daniel Fernex, B. R. Noack, R. Semaan            2020-10-30 arXiv: Data Analysis, Statistics and Probability 1 49 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba/","title":"455bfc515eb279cc09023faa1f78c6efb61224ba","text":""},{"location":"recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba/#_1","title":"455bfc515eb279cc09023faa1f78c6efb61224ba","text":"<p> This page was last updated on 2024-09-11 16:11:09 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Graph-Guided Network for Irregularly Sampled Multivariate Time Series Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off PrimeNet: Pre-training for Irregular Multivariate Time Series              Ranak Roy Chowdhury, Jiacheng Li, Xiyuan Zhang, Dezhi Hong, Rajesh K. Gupta, Jingbo Shang            2023-06-26 DBLP 9 16 visibility_off Finding Short Signals in Long Irregular Time Series with Continuous-Time Attention Policy Networks              Thomas Hartvigsen, Jidapa Thadajarassiri, Xiangnan Kong, Elke A. Rundensteiner            2023-02-08 ArXiv 1 38 visibility_off Continuous Time Evidential Distributions for Irregular Time Series              Taylor W. Killian, Haoran Zhang, Thomas Hartvigsen, Ava P. Amini            2023-07-25 ArXiv 0 12 visibility_off Time Series as Images: Vision Transformer for Irregularly Sampled Time Series              Zekun Li, SHIYANG LI, Xifeng Yan            2023-03-01 ArXiv 12 65 visibility_off Time Series as Images: Vision Transformer for Irregularly Sampled Time Series              Zekun Li, SHIYANG LI, Xifeng Yan            2023-03-01 ArXiv 12 65 visibility_off Compatible Transformer for Irregularly Sampled Multivariate Time Series              Yuxi Wei, Juntong Peng, Tong He, Chenxin Xu, Jian Zhang, Shirui Pan, Siheng Chen            2023-10-17 2023 IEEE International Conference on Data Mining (ICDM) 0 9 visibility_off Learning from Irregularly-Sampled Time Series: A Missing Data Perspective              Steven Cheng-Xian Li, Benjamin M Marlin            2020-07-12 ArXiv, DBLP, MAG 48 33 visibility_off Continual Learning for Multivariate Time Series Tasks with Variable Input Dimensions              Vibhor Gupta, Jyoti Narwariya, Pankaj Malhotra, L. Vig, Gautam M. Shroff            2021-12-01 2021 IEEE International Conference on Data Mining (ICDM) 16 27 visibility_off TimeAutoML: Autonomous Representation Learning for Multivariate Irregularly Sampled Time Series              Yang Jiao, Kai Yang, Shaoyu Dou, Pan Luo, Sijia Liu, Dongjin Song            2020-10-04 ArXiv 6 22 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01/","title":"4971f9abd024e40fbbdff2e9492745b68a6bca01","text":""},{"location":"recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01/#_1","title":"4971f9abd024e40fbbdff2e9492745b68a6bca01","text":"<p> This page was last updated on 2024-09-11 16:12:03 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Sparse identification of Lagrangian for nonlinear dynamical systems via proximal gradient method              Adam Purnomo, M. Hayashibe            2022-09-04 Scientific Reports 2 23 visibility_off Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI)              Gabriel F. Machado, Morgan Jones            2023-10-06 2024 American Control Conference (ACC) 1 1 visibility_off PySINDy: A comprehensive Python package for robust sparse system identification              A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2021-11-12 J. Open Source Softw. 111 65 visibility_off Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian              H. Chu, M. Hayashibe            2020-01-31 IEEE Robotics and Automation Letters 23 23 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 8 2 visibility_off Derivative-Based SINDy (DSINDy): Addressing the Challenge of Discovering Governing Equations from Noisy Data              J. Wentz, A. Doostan            2022-11-10 SSRN Electronic Journal 10 33 visibility_off A Robust SINDy Approach by Combining Neural Networks and an Integral Form              A. Forootani, P. Goyal, P. Benner            2023-09-13 ArXiv 2 14 visibility_off Generalizing the SINDy approach with nested neural networks              Camilla Fiorini, Cl'ement Flint, Louis Fostier, Emmanuel Franck, Reyhaneh Hashemi, Victor Michel-Dansac, Wassim Tenachi            2024-04-24 ArXiv 0 3 visibility_off SINDy vs Hard Nonlinearities and Hidden Dynamics: a Benchmarking Study              Aurelio Raffa Ugolini, Valentina Breschi, Andrea Manzoni, M. Tanelli            2024-03-01 ArXiv 1 1 visibility_off Weak SINDy For Partial Differential Equations              D. Messenger, D. Bortz            2020-07-06 Journal of computational physics 126 20 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/4a111f7a3b56d0468f13104999844885157ef17d/","title":"4a111f7a3b56d0468f13104999844885157ef17d","text":""},{"location":"recommendations/4a111f7a3b56d0468f13104999844885157ef17d/#_1","title":"4a111f7a3b56d0468f13104999844885157ef17d","text":"<p> This page was last updated on 2024-09-11 16:11:24 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Unified Training of Universal Time Series Forecasting Transformers Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Chronos: Learning the Language of Time Series              Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang            2024-03-12 ArXiv 28 18 visibility_off Timer: Generative Pre-trained Transformers Are Large Time Series Models              Yong Liu, Haoran Zhang, Chenyu Li, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-02-04 ArXiv, DBLP 4 66 visibility_off A Time Series is Worth 64 Words: Long-term Forecasting with Transformers              Yuqi Nie, Nam H. Nguyen, Phanwadee Sinthong, J. Kalagnanam            2022-11-27 ArXiv 534 34 visibility_off HiMTM: Hierarchical Multi-Scale Masked Time Series Modeling with Self-Distillation for Long-Term Forecasting              Shubao Zhao, Ming Jin, Zhaoxiang Hou, Che-Sheng Yang, Zengxiang Li, Qingsong Wen, Yi Wang            2024-01-10 ArXiv 0 6 visibility_off Two Steps Forward and One Behind: Rethinking Time Series Forecasting with Deep Learning              Riccardo Ughi, Eugenio Lomurno, Matteo Matteucci            2023-04-10 ArXiv, DBLP 1 6 visibility_off Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain              Gerald Woo, Chenghao Liu, Akshat Kumar, Doyen Sahoo            2023-10-08 ArXiv 7 22 visibility_off FAITH: Frequency-domain Attention In Two Horizons for Time Series Forecasting              Ruiqi Li, Maowei Jiang, Kai Wang, Kaiduo Feng, Quangao Liu, Yue Sun, Xiufang Zhou            2024-05-22 ArXiv 0 2 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14/","title":"5bd2c0acaf58c25f71617db2396188c74d29bf14","text":""},{"location":"recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14/#_1","title":"5bd2c0acaf58c25f71617db2396188c74d29bf14","text":"<p> This page was last updated on 2024-09-11 16:11:32 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Domain Adaptation for Time Series Under Feature and Label Shifts Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Deep Unsupervised Domain Adaptation for Time Series Classification: a Benchmark              Hassan Ismail Fawaz, Ganesh Del Grosso, Tanguy Kerdoncuff, Aur\u00e9lie Boisbunon, Illyyne Saffar            2023-12-15 ArXiv 1 5 visibility_off A Virtual-Label-Based Hierarchical Domain Adaptation Method for Time-Series Classification.              Wenmian Yang, Lizhi Cheng, Mohamed Ragab, Min Wu, Sinno Jialin Pan, Zhenghua Chen            2024-08-28 IEEE transactions on neural networks and learning systems 0 14 visibility_off ADATIME: A Benchmarking Suite for Domain Adaptation on Time Series Data              Mohamed Ragab, Emadeldeen Eldele, Wee Ling Tan, Chuan-Sheng Foo, Zhenghua Chen, Min Wu, C. Kwoh, Xiaoli Li            2022-03-15 ACM Transactions on Knowledge Discovery from Data 40 37 visibility_off Domain Generalization via Selective Consistency Regularization for Time Series Classification              Wenyu Zhang, Mohamed Ragab, Chuan-Sheng Foo            2022-06-16 2022 26th International Conference on Pattern Recognition (ICPR) 0 25 visibility_off Domain Adaptation for Time Series Forecasting via Attention Sharing              Xiaoyong Jin, Youngsuk Park, Danielle C. Maddix, Bernie Wang, Xifeng Yan            2021-02-13 ArXiv, DBLP 57 65 visibility_off Domain Adaptation with Representation Learning and Nonlinear Relation for Time Series              A. Hussein, Hazem Hajj            2022-02-15 ACM Transactions on Internet of Things 11 9 visibility_off Match-And-Deform: Time Series Domain Adaptation through Optimal Transport and Temporal Alignment              Franccois Painblanc, L. Chapel, N. Courty, Chlo\u00e9 Friguet, Charlotte Pelletier, R. Tavenard            2023-08-24 ArXiv 2 34 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/5d150cec2775f9bc863760448f14104cc8f42368/","title":"5d150cec2775f9bc863760448f14104cc8f42368","text":""},{"location":"recommendations/5d150cec2775f9bc863760448f14104cc8f42368/#_1","title":"5d150cec2775f9bc863760448f14104cc8f42368","text":"<p> This page was last updated on 2024-09-11 16:11:54 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Discovering governing equations from data by sparse identification of nonlinear dynamical systems Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Data-driven discovery of coordinates and governing equations              Kathleen P. Champion, Bethany Lusch, J. Kutz, S. Brunton            2019-03-29 Proceedings of the National Academy of Sciences of the United States of America 615 65 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 8 2 visibility_off Sparse Estimation for Hamiltonian Mechanics              Yuya Note, Masahito Watanabe, Hiroaki Yoshimura, Takaharu Yaguchi, Toshiaki Omori            2024-03-25 Mathematics 0 11 visibility_off Physics-informed learning of governing equations from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 Nature Communications 250 12 visibility_off Sparsistent Model Discovery              Georges Tod, G. Both, R. Kusters            2021-06-22 ArXiv 1 12 visibility_off Deep learning of physical laws from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 ArXiv 19 12 visibility_off Discovery of nonlinear dynamical systems using a Runge\u2013Kutta inspired dictionary-based sparse regression approach              P. Goyal, P. Benner            2021-05-11 Proceedings. Mathematical, Physical, and Engineering Sciences 38 53 visibility_off Exploiting sparsity and equation-free architectures in complex systems              J. Proctor, S. Brunton, Bingni W. Brunton, J. Kutz            2014-12-10 The European Physical Journal Special Topics 64 65 visibility_off Exploiting sparsity and equation-free architectures in complex systems              J. Proctor, S. Brunton, Bingni W. Brunton, J. Kutz            2014-12-01 The European Physical Journal Special Topics 7 65 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c/","title":"60d0d998fa038182b3b69a57adb9b2f82d40589c","text":""},{"location":"recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c/#_1","title":"60d0d998fa038182b3b69a57adb9b2f82d40589c","text":"<p> This page was last updated on 2024-09-11 16:11:54 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Robust learning from noisy, incomplete, high-dimensional experimental data via physically constrained symbolic regression Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Automated discovery of fundamental variables hidden in experimental data              Boyuan Chen, Kuang Huang, Sunand Raghupathi, I. Chandratreya, Qi Du, H. Lipson            2022-07-01 Nature Computational Science 69 20 visibility_off Opportunities for machine learning in scientific discovery              Ricardo Vinuesa, Jean Rabault, Hossein Azizpour, Stefan Bauer, Bingni W. Brunton, Arne Elofsson, Elias Jarlebring, Hedvig Kjellstrom, Stefano Markidis, David Marlevi, Paola Cinnella, S. Brunton            2024-05-07 ArXiv 1 65 visibility_off DeepMoD: Deep learning for model discovery in noisy data              G. Both, Subham Choudhury, P. Sens, R. Kusters            2019-04-20 J. Comput. Phys. 101 34 visibility_off Discovering sparse interpretable dynamics from partial observations              Peter Y. Lu, Joan Ari\u00f1o Bernad, M. Solja\u010di\u0107            2021-07-22 Communications Physics 17 94 visibility_off PNAS Plus Significance Statements              Ronald R. Coifman, David A. Kessler, A. Goodkind            2017-09-19 Proceedings of the National Academy of Sciences 40 13 visibility_off Discovering State Variables Hidden in Experimental Data              Boyuan Chen, Kuang Huang, Sunand Raghupathi, I. Chandratreya, Qi Du, Hod Lipson            2021-12-20 ArXiv 13 73 visibility_off A physics-informed operator regression framework for extracting data-driven continuum models              Ravi G. Patel, N. Trask, M. Wood, E. Cyr            2020-09-25 ArXiv 88 17 visibility_off Data-Driven Discovery of Coarse-Grained Equations              Joseph Bakarji, D. Tartakovsky            2020-01-30 J. Comput. Phys. 32 47 visibility_off Universal Differential Equations for Scientific Machine Learning              Christopher Rackauckas, Yingbo Ma, Julius Martensen, Collin Warner, K. Zubov, R. Supekar, Dominic J. Skinner, A. Ramadhan            2020-01-13 ArXiv 493 25 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37/","title":"648d90b713997a771e2c49f02cd771e8b7b10b37","text":""},{"location":"recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37/#_1","title":"648d90b713997a771e2c49f02cd771e8b7b10b37","text":"<p> This page was last updated on 2024-09-11 16:11:32 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistency Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Self-Supervised Contrastive Learning for Medical Time Series: A Systematic Review              Ziyu Liu, A. Alavi, Minyi Li, X. Zhang            2023-04-23 Sensors (Basel, Switzerland) 22 77 visibility_off Time-series representation learning via Time-Frequency Fusion Contrasting              Wenbo Zhao, Ling Fan            2024-06-12 Frontiers in Artificial Intelligence 0 1 visibility_off TS-MoCo: Time-Series Momentum Contrast for Self-Supervised Physiological Representation Learning              Philipp Hallgarten, David Bethge, Ozan \u00d6zdenizci, T. Gro\u00dfe-Puppendahl, Enkelejda Kasneci            2023-06-10 2023 31st European Signal Processing Conference (EUSIPCO) 0 38 visibility_off Supervised Contrastive Few-Shot Learning for High-Frequency Time Series              X. Chen, Cheng Ge, Ming Wang, Jin Wang            2023-06-26 DBLP 2 140 visibility_off Contrast Everything: A Hierarchical Contrastive Framework for Medical Time-Series              Yihe Wang, Yu Han, Haishuai Wang, Xiang Zhang            2023-10-21 ArXiv 21 3 visibility_off CALDA: Improving Multi-Source Time Series Domain Adaptation With Contrastive Adversarial Learning              Garrett Wilson, J. Doppa, D. Cook            2021-09-30 IEEE Transactions on Pattern Analysis and Machine Intelligence 16 72 visibility_off Self-supervised Classification of Clinical Multivariate Time Series using Time Series Dynamics              Yakir Yehuda, Daniel Freedman, Kira Radinsky            2023-08-04 Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 0 24 visibility_off Phase-driven Domain Generalizable Learning for Nonstationary Time Series              Payal Mohapatra, Lixu Wang, Qi Zhu            2024-02-05 ArXiv 1 9 visibility_off Contrastive Neural Processes for Self-Supervised Learning              Konstantinos Kallidromitis, Denis A. Gudovskiy, Kozuka Kazuki, Ohama Iku, Luca Rigazio            2021-10-24 ArXiv, DBLP 10 13 visibility_off Large Scale Time-Series Representation Learning via Simultaneous Low and High Frequency Feature Bootstrapping              Vandan Gorade, Azad Singh, Deepak Mishra            2022-04-24 IEEE transactions on neural networks and learning systems 5 6 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/73dd9c49f205280991826b2ea4b50344203916b4/","title":"73dd9c49f205280991826b2ea4b50344203916b4","text":""},{"location":"recommendations/73dd9c49f205280991826b2ea4b50344203916b4/#_1","title":"73dd9c49f205280991826b2ea4b50344203916b4","text":"<p> This page was last updated on 2024-09-11 16:12:04 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Learning Discrepancy Models From Experimental Data Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Sparse identification of Lagrangian for nonlinear dynamical systems via proximal gradient method              Adam Purnomo, M. Hayashibe            2022-09-04 Scientific Reports 2 23 visibility_off Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian              H. Chu, M. Hayashibe            2020-01-31 IEEE Robotics and Automation Letters 23 23 visibility_off SINDy vs Hard Nonlinearities and Hidden Dynamics: a Benchmarking Study              Aurelio Raffa Ugolini, Valentina Breschi, Andrea Manzoni, M. Tanelli            2024-03-01 ArXiv 1 1 visibility_off Machine Learning and System Identification for Estimation in Physical Systems              Fredrik Bagge Carlson            2018-12-20 ArXiv 5 8 visibility_off Composed Physics- and Data-driven System Identification for Non-autonomous Systems in Control Engineering              Ricarda-Samantha G\u00f6tte, Julia Timmermann            2021-12-15 2022 3rd International Conference on Artificial Intelligence, Robotics and Control (AIRC) 4 5 visibility_off Sparse identification of nonlinear dynamics for model predictive control in the low-data limit              E. Kaiser, J. Kutz, S. Brunton            2017-11-15 Proceedings. Mathematical, Physical, and Engineering Sciences 435 65 visibility_off Data-based modeling and control of nonlinear process systems using sparse identification: An overview of recent results              Fahim Abdullah, P. Christofides            2023-03-01 Comput. Chem. Eng. 12 76 visibility_off SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics              Kadierdan Kaheman, J. Kutz, S. Brunton            2020-04-05 Proceedings. Mathematical, Physical, and Engineering Sciences 202 65 visibility_off Learning Dynamical Systems by Leveraging Data from Similar Systems              Lei Xin, Lintao Ye, G. Chiu, S. Sundaram            2023-02-08 ArXiv 7 36 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/883547fdbd88552328a6615ec620f96e39c57018/","title":"883547fdbd88552328a6615ec620f96e39c57018","text":""},{"location":"recommendations/883547fdbd88552328a6615ec620f96e39c57018/#_1","title":"883547fdbd88552328a6615ec620f96e39c57018","text":"<p> This page was last updated on 2024-09-11 16:12:07 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Ensemble-SINDy: Robust sparse model discovery in the low-data, high-noise limit, with active learning and control Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Sparsifying priors for Bayesian uncertainty quantification in model discovery              Seth M. Hirsh, D. Barajas-Solano, J. Kutz            2021-07-05 Royal Society Open Science 54 31 visibility_off Convergence of uncertainty estimates in Ensemble and Bayesian sparse model discovery              Liyao (Mars) Gao, Urban Fasel, S. Brunton, J. Kutz            2023-01-30 ArXiv 11 65 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 8 2 visibility_off SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics              Kadierdan Kaheman, J. Kutz, S. Brunton            2020-04-05 Proceedings. Mathematical, Physical, and Engineering Sciences 202 65 visibility_off Rapid Bayesian identification of sparse nonlinear dynamics from scarce and noisy data              Lloyd Fung, Urban Fasel, M. Juniper            2024-02-23 ArXiv 1 38 visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3256 65 visibility_off Sparse identification of nonlinear dynamics in the presence of library and system uncertainty              Andrew O'Brien            2024-01-23 ArXiv 0 0 $\\dot { \\boldsymbol x} = { \\boldsymbol f} ({ \\boldsymbol x})$ . First, we propose, for use in high-noise settings, an extensive toolkit of critically enabling extensions for the SINDy regression method, to progressively cull functionals from an over-complete library and yield a set of sparse equations that regress to the derivate  $\\dot { \\boldsymbol {x}}$ . This toolkit includes: (regression step) weight timepoints based on estimated noise, use ensembles to estimate coefficients, and regress using FFTs; (culling step) leverage linear dependence of functionals, and restore and protect culled functionals based on Figures of Merit (FoMs). In a novel Assessment step, we define FoMs that compare model predictions to the original time-series (i.e.,  ${ \\boldsymbol x}(t)$  rather than  $\\dot { \\boldsymbol {x}}(t)$ ). These innovations can extract sparse governing equations and coefficients from high-noise time-series data (e.g., 300% added noise). For example, it discovers the correct sparse libraries in the Lorenz system, with median coefficient estimate errors equal to 1%\u22123% (for 50% noise), 6%\u22128% (for 100% noise), and 23%\u221225% (for 300% noise). The enabling modules in the toolkit are combined into a single method, but the individual modules can be tactically applied in other equation discovery methods (SINDy or not) to improve results on high-noise data. Second, we propose a technique, applicable to any model discovery method based on  $\\dot { \\boldsymbol x} = { \\boldsymbol f} ({ \\boldsymbol x})$ , to assess the accuracy of a discovered model in the context of non-unique solutions due to noisy data. Currently, this non-uniqueness can obscure a discovered model\u2019s accuracy and thus a discovery method\u2019s effectiveness. We describe a technique that uses linear dependencies among functionals to transform a discovered model into an equivalent form that is closest to the true model, enabling more accurate assessment of a discovered model\u2019s correctness.\"&gt;           visibility_off A Toolkit for Data-Driven Discovery of Governing Equations in High-Noise Regimes              Charles B. Delahunt, J. Kutz            2021-11-08 IEEE Access 16 31 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/b2eb064f432557c59ce99834d7dc7817e4687271/","title":"B2eb064f432557c59ce99834d7dc7817e4687271","text":""},{"location":"recommendations/b2eb064f432557c59ce99834d7dc7817e4687271/#_1","title":"B2eb064f432557c59ce99834d7dc7817e4687271","text":"<p> This page was last updated on 2024-09-11 16:11:57 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Sparse identification of nonlinear dynamics for model predictive control in the low-data limit Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off SINDy with Control: A Tutorial              Urban Fasel, E. Kaiser, J. Kutz, Bingni W. Brunton, S. Brunton            2021-08-30 2021 60th IEEE Conference on Decision and Control (CDC) 48 65 visibility_off SINDy vs Hard Nonlinearities and Hidden Dynamics: a Benchmarking Study              Aurelio Raffa Ugolini, Valentina Breschi, Andrea Manzoni, M. Tanelli            2024-03-01 ArXiv 1 1 visibility_off Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian              H. Chu, M. Hayashibe            2020-01-31 IEEE Robotics and Automation Letters 23 23 visibility_off SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics              Kadierdan Kaheman, J. Kutz, S. Brunton            2020-04-05 Proceedings. Mathematical, Physical, and Engineering Sciences 202 65 visibility_off Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI)              Gabriel F. Machado, Morgan Jones            2023-10-06 2024 American Control Conference (ACC) 1 1 visibility_off PySINDy: A comprehensive Python package for robust sparse system identification              A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2021-11-12 J. Open Source Softw. 111 65 visibility_off Discovering sparse interpretable dynamics from partial observations              Peter Y. Lu, Joan Ari\u00f1o Bernad, M. Solja\u010di\u0107            2021-07-22 Communications Physics 17 94 visibility_off PySINDy: A Python package for the sparse identification of nonlinear dynamical systems from data              Brian M. de Silva, Kathleen P. Champion, M. Quade, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2020-05-18 J. Open Source Softw. 129 65 visibility_off Learning Linear Representations of Nonlinear Dynamics Using Deep Learning              Akhil Ahmed, E. A. Rio-Chanona, Mehmet Mercang\u00f6z            2022-04-03 ArXiv 3 19 visibility_off Data-based modeling and control of nonlinear process systems using sparse identification: An overview of recent results              Fahim Abdullah, P. Christofides            2023-03-01 Comput. Chem. Eng. 12 76 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da/","title":"C0fc3882a9976f6a9cdc3a724bce184b786503da","text":""},{"location":"recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da/#_1","title":"C0fc3882a9976f6a9cdc3a724bce184b786503da","text":"<p> This page was last updated on 2024-09-11 16:12:08 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article A Unified Framework for Sparse Relaxed Regularized Regression: SR3 Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Sparse Relaxed Regularized Regression: SR3              P. Zheng, T. Askham, S. Brunton, J. Kutz, A. Aravkin            2018-07-14 ArXiv 9 65 visibility_off Rank-one Convexification for Sparse Regression              Alper Atamt\u00fcrk, A. G\u00f3mez            2019-01-29 ArXiv 49 35 visibility_off Sparse Recovery via Partial Regularization: Models, Theory and Algorithms              Zhaosong Lu, Xiaorui Li            2015-11-23 ArXiv 36 32 visibility_off Structured Regularizers for High-Dimensional Problems: Statistical and Computational Issues              M. Wainwright            2014-01-03 59 95 visibility_off Compressed Sparse Linear Regression              S. Kasiviswanathan, M. Rudelson            2017-07-25 ArXiv 1 30 visibility_off WARPd: A linearly convergent first-order method for inverse problems with approximate sharpness conditions              Matthew J. Colbrook            2021-10-24 ArXiv 2 16 $\\ell _1$ -norm as the loss function for the residual error and utilizes a generalized nonconvex penalty for sparsity inducing. The $\\ell _1$ -loss is less sensitive to outliers in the measurements than the popular  $\\ell _2$-loss, while the nonconvex penalty has the capability of ameliorating the bias problem of the popular convex LASSO penalty and thus can yield more accurate recovery. To solve this nonconvex and nonsmooth minimization formulation efficiently, we propose a first-order algorithm based on alternating direction method of multipliers. A smoothing strategy on the $\\ell _1$ -loss function has been used in deriving the new algorithm to make it convergent. Further, a sufficient condition for the convergence of the new algorithm has been provided for generalized nonconvex regularization. In comparison with several state-of-the-art algorithms, the new algorithm showed better performance in numerical experiments in recovering sparse signals and compressible images. The new algorithm scales well for large-scale problems, as often encountered in image processing.\"&gt;           visibility_off Efficient and Robust Recovery of Sparse Signal and Image Using Generalized Nonconvex Regularization              Fei Wen, L. Pei, Yuan Yang, Wenxian Yu, Peilin Liu            2017-03-23 IEEE Transactions on Computational Imaging 85 30 visibility_off Regularizers for structured sparsity              C. Micchelli, Jean Morales, M. Pontil            2010-10-04 Advances in Computational Mathematics 80 70 visibility_off Regularizers for structured sparsity              C. Micchelli, Jean Morales, M. Pontil            2010-10-04 Advances in Computational Mathematics 80 70 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524/","title":"C3c94ccc094dcf546e8e31c9a42506302e837524","text":""},{"location":"recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524/#_1","title":"C3c94ccc094dcf546e8e31c9a42506302e837524","text":"<p> This page was last updated on 2024-09-11 16:11:33 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article AZ-whiteness test: a test for signal uncorrelation on spatio-temporal graphs Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off AZ-whiteness test: a test for uncorrelated noise on spatio-temporal graphs              Daniele Zambon, C. Alippi            2022-04-23 ArXiv 5 50 visibility_off A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection              Ming Jin, Huan Yee Koh, Qingsong Wen, Daniele Zambon, C. Alippi, G. I. Webb, Irwin King, Shirui Pan            2023-07-07 IEEE transactions on pattern analysis and machine intelligence 64 50 visibility_off Learning Time-Aware Graph Structures for Spatially Correlated Time Series Forecasting              Minbo Ma, Jilin Hu, Christian S. Jensen, Fei Teng, Peng Han, Zhiqiang Xu, Tian-Jie Li            2023-12-27 2024 IEEE 40th International Conference on Data Engineering (ICDE) 0 12 visibility_off Graph construction on complex spatiotemporal data for enhancing graph neural network-based approaches              Stefan Bloemheuvel, J. Hoogen, Martin Atzmueller            2023-09-25 International Journal of Data Science and Analytics 0 6 visibility_off Time-Varying Signals Recovery Via Graph Neural Networks              Jhon A. Castro-Correa, Jhony H. Giraldo, Anindya Mondal, M. Badiey, T. Bouwmans, Fragkiskos D. Malliaros            2023-02-22 ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 5 42 visibility_off Graph-Time Convolutional Neural Networks: Architecture and Theoretical Analysis              Mohammad Sabbaqi, E. Isufi            2022-06-30 IEEE Transactions on Pattern Analysis and Machine Intelligence 7 21 visibility_off Multivariate Time Series Forecasting With Dynamic Graph Neural ODEs              Ming Jin, Yu Zheng, Yuanhao Li, Siheng Chen, B. Yang, Shirui Pan            2022-02-17 IEEE Transactions on Knowledge and Data Engineering 60 43 visibility_off Dynamic Graph Learning with Long and Short-Term for Multivariate Time Series Anomaly Detection              Yuyin Tian, Rong Gao, Lingyu Yan, Donghua Liu, Zhiwei Ye            2023-09-07 2023 IEEE 12th International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS) 0 4 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/ccea298edb788edf821aef58f0952c3e8debc25a/","title":"Ccea298edb788edf821aef58f0952c3e8debc25a","text":""},{"location":"recommendations/ccea298edb788edf821aef58f0952c3e8debc25a/#_1","title":"Ccea298edb788edf821aef58f0952c3e8debc25a","text":"<p> This page was last updated on 2024-09-11 16:11:16 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Graph Deep Learning for Time Series Forecasting Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting              Nancy R. Xu, Chrysoula Kosma, M. Vazirgiannis            2023-07-27 ArXiv 0 54 visibility_off Balanced Graph Structure Learning for Multivariate Time Series Forecasting              Weijun Chen, Yanze Wang, Chengshuo Du, Zhenglong Jia, Feng Liu, Ran Chen            2022-01-24 ArXiv 0 10 visibility_off Sparse Graph Learning from Spatiotemporal Time Series              Andrea Cini, Daniele Zambon, C. Alippi            2022-05-26 J. Mach. Learn. Res. 11 50 visibility_off ForecastGrapher: Redefining Multivariate Time Series Forecasting with Graph Neural Networks              Wanlin Cai, Kun Wang, Hao Wu, Xiaoxu Chen, Yuankai Wu            2024-05-28 ArXiv 0 3 visibility_off Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks              Zonghan Wu, Shirui Pan, Guodong Long, Jing Jiang, Xiaojun Chang, Chengqi Zhang            2020-05-24 Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery &amp; Data Mining 981 55 visibility_off DeepHGNN: Study of Graph Neural Network based Forecasting Methods for Hierarchically Related Multivariate Time Series              Abishek Sriramulu, Nicolas Fourrier, Christoph Bergmeir            2024-05-29 ArXiv 0 4 visibility_off FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective              Kun Yi, Qi Zhang, Wei Fan, Hui He, Liang Hu, Pengyang Wang, Ning An, Longbin Cao, Zhendong Niu            2023-11-10 ArXiv 32 6 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9/","title":"D3dbbd0f0de51b421a6220bd6480b8d2e99a88e9","text":""},{"location":"recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9/#_1","title":"D3dbbd0f0de51b421a6220bd6480b8d2e99a88e9","text":"<p> This page was last updated on 2024-09-11 16:11:08 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Graph Time-series Modeling in Deep Learning: A Survey              Hongjie Chen, Hoda Eldardiry            2023-12-23 ACM Transactions on Knowledge Discovery from Data 1 15 visibility_off TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting              Nancy R. Xu, Chrysoula Kosma, M. Vazirgiannis            2023-07-27 ArXiv 0 54 visibility_off Graph Deep Learning for Time Series Forecasting              Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-10-24 ArXiv 6 50 visibility_off Graph Anomaly Detection in Time Series: A Survey              Thi Kieu Khanh Ho, Ali Karami, N. Armanfard            2023-01-31 ArXiv 3 15 visibility_off Multivariate Time-Series Anomaly Detection based on Enhancing Graph Attention Networks with Topological Analysis              Zhe Liu, Xiang Huang, Jingyun Zhang, Zhifeng Hao, L. Sun, Hao Peng            2024-08-23 ArXiv 0 2 visibility_off FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective              Kun Yi, Qi Zhang, Wei Fan, Hui He, Liang Hu, Pengyang Wang, Ning An, Longbin Cao, Zhendong Niu            2023-11-10 ArXiv 32 6 visibility_off MGADN: A Multi-task Graph Anomaly Detection Network for Multivariate Time Series              Wei-Shu Xiong, Xiaochen (Michael) Sun            2022-11-22 ArXiv 1 2 visibility_off Multivariate Time Series Anomaly Detection via Dynamic Graph Forecasting              Katrina Chen, M. Feng, T. Wirjanto            2023-02-04 ArXiv 4 22 visibility_off Edge Conditional Node Update Graph Neural Network for Multi-variate Time Series Anomaly Detection              H. Jo, Seong-Whan Lee            2024-01-25 Inf. Sci. 0 1 visibility_off Learning Time-Aware Graph Structures for Spatially Correlated Time Series Forecasting              Minbo Ma, Jilin Hu, Christian S. Jensen, Fei Teng, Peng Han, Zhiqiang Xu, Tian-Jie Li            2023-12-27 2024 IEEE 40th International Conference on Data Engineering (ICDE) 0 12 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419/","title":"E2a83369383aff37224170c1ae3d3870d5d9e419","text":""},{"location":"recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419/#_1","title":"E2a83369383aff37224170c1ae3d3870d5d9e419","text":"<p> This page was last updated on 2024-09-11 16:11:15 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Taming Local Effects in Graph-based Spatiotemporal Forecasting Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Graph Deep Learning for Time Series Forecasting              Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-10-24 ArXiv 6 50 visibility_off Scalable Spatiotemporal Graph Neural Networks              Andrea Cini, Ivan Marisca, F. Bianchi, C. Alippi            2022-09-14 ArXiv 30 50 visibility_off Sparse Graph Learning from Spatiotemporal Time Series              Andrea Cini, Daniele Zambon, C. Alippi            2022-05-26 J. Mach. Learn. Res. 11 50 visibility_off Unified Spatio-Temporal Graph Neural Networks: Data-Driven Modeling for Social Science              Yifan Li, Yu Lin, Y. Gao, L. Khan            2022-07-18 2022 International Joint Conference on Neural Networks (IJCNN) 0 12 visibility_off ForecastGrapher: Redefining Multivariate Time Series Forecasting with Graph Neural Networks              Wanlin Cai, Kun Wang, Hao Wu, Xiaoxu Chen, Yuankai Wu            2024-05-28 ArXiv 0 3 visibility_off FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective              Kun Yi, Qi Zhang, Wei Fan, Hui He, Liang Hu, Pengyang Wang, Ning An, Longbin Cao, Zhendong Niu            2023-11-10 ArXiv 32 6 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6/","title":"E2e1f1b8e6c1b7f4f166e15b7c674945856a51b6","text":""},{"location":"recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6/#_1","title":"E2e1f1b8e6c1b7f4f166e15b7c674945856a51b6","text":"<p> This page was last updated on 2024-09-11 16:11:31 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters              Mouxiang Chen, Lefei Shen, Zhuo Li, Xiaoyun Joy Wang, Jianling Sun, Chenghao Liu            2024-08-30 ArXiv 0 3 visibility_off Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction              Cheng Feng, Long Huang, Denis Krompass            2024-02-12 ArXiv 3 14 visibility_off Generative Pre-Trained Diffusion Paradigm for Zero-Shot Time Series Forecasting              Jiarui Yang, Tao Dai, Naiqi Li, Junxi Wu, Peiyuan Liu, Jinmin Li, Jigang Bao, Haigang Zhang, Shu-Tao Xia            2024-06-04 ArXiv 0 4 visibility_off Chronos: Learning the Language of Time Series              Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang            2024-03-12 ArXiv 28 18 visibility_off Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting              Kashif Rasul, Arjun Ashok, Andrew Robert Williams, Arian Khorasani, George Adamopoulos, Rishika Bhagwatkar, Marin Bilovs, Hena Ghonia, N. Hassen, Anderson Schneider, Sahil Garg, Alexandre Drouin, Nicolas Chapados, Yuriy Nevmyvaka, I. Rish            2023-10-12 ArXiv 16 40 visibility_off A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model              Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung            2024-05-03 ArXiv 4 47 visibility_off Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learning              Yuxuan Bian, Xu Ju, Jiangtong Li, Zhijian Xu, Dawei Cheng, Qiang Xu            2024-02-07 ArXiv 5 5 visibility_off Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domain              Gerald Woo, Chenghao Liu, Akshat Kumar, Doyen Sahoo            2023-10-08 ArXiv 7 22 visibility_off One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors              Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin            2023-11-24 ArXiv 4 7 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8/","title":"E6f0a85009481dcfd93aaa43ed3f980e5033b0d8","text":""},{"location":"recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8/#_1","title":"E6f0a85009481dcfd93aaa43ed3f980e5033b0d8","text":"<p> This page was last updated on 2024-09-11 16:12:07 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Learning sparse nonlinear dynamics via mixed-integer optimization Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 8 2 visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3256 65 visibility_off Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI)              Gabriel F. Machado, Morgan Jones            2023-10-06 2024 American Control Conference (ACC) 1 1 visibility_off PySINDy: A comprehensive Python package for robust sparse system identification              A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2021-11-12 J. Open Source Softw. 111 65 visibility_off Sparse reconstruction of ordinary differential equations with inference              S. Venkatraman, Sumanta Basu, M. Wells            2023-08-17 ArXiv 0 40 visibility_off Physics-informed learning of governing equations from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 Nature Communications 250 12 visibility_off PySINDy: A Python package for the Sparse Identification of Nonlinear Dynamics from Data.              Brian M. de Silva, Kathleen P. Champion, M. Quade, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2020-04-17 arXiv: Dynamical Systems 49 65 visibility_off Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian              H. Chu, M. Hayashibe            2020-01-31 IEEE Robotics and Automation Letters 23 23 visibility_off Sparse learning of stochastic dynamical equations.              L. Boninsegna, F. N\u00fcske, C. Clementi            2017-12-06 The Journal of chemical physics 193 45 visibility_off Sparsistent Model Discovery              Georges Tod, G. Both, R. Kusters            2021-06-22 ArXiv 1 12 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f/","title":"F45f85fa1beaa795c24c4ff86f1f2deece72252f","text":""},{"location":"recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f/#_1","title":"F45f85fa1beaa795c24c4ff86f1f2deece72252f","text":"<p> This page was last updated on 2024-09-11 16:11:18 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article A decoder-only foundation model for time-series forecasting Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting              Kashif Rasul, Arjun Ashok, Andrew Robert Williams, Arian Khorasani, George Adamopoulos, Rishika Bhagwatkar, Marin Bilovs, Hena Ghonia, N. Hassen, Anderson Schneider, Sahil Garg, Alexandre Drouin, Nicolas Chapados, Yuriy Nevmyvaka, I. Rish            2023-10-12 ArXiv 16 40 visibility_off Are Language Models Actually Useful for Time Series Forecasting?              Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen            2024-06-22 ArXiv 1 1 visibility_off Time-LLM: Time Series Forecasting by Reprogramming Large Language Models              Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen            2023-10-03 ArXiv 134 9 visibility_off In-context Time Series Predictor              Jiecheng Lu, Yan Sun, Shihao Yang            2024-05-23 ArXiv 0 1 visibility_off LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters              Ching Chang, Wenjie Peng, Tien-Fu Chen            2023-08-16 ArXiv 16 2 visibility_off A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model              Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung            2024-05-03 ArXiv 4 47 visibility_off LSTPrompt: Large Language Models as Zero-Shot Time Series Forecasters by Long-Short-Term Prompting              Haoxin Liu, Zhiyuan Zhao, Jindong Wang, Harshavardhan Kamarthi, B. A. Prakash            2024-02-25 ArXiv 6 7 visibility_off Time-FFM: Towards LM-Empowered Federated Foundation Model for Time Series Forecasting              Qingxiang Liu, Xu Liu, Chenghao Liu, Qingsong Wen, Yuxuan Liang            2024-05-23 ArXiv 0 6 visibility_off One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors              Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin            2023-11-24 ArXiv 4 7 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90/","title":"Ff1f1cf9df8c413ec7345da7604ba28597da5b90","text":""},{"location":"recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90/#_1","title":"Ff1f1cf9df8c413ec7345da7604ba28597da5b90","text":"<p> This page was last updated on 2024-09-11 16:11:34 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article UNITS: A Unified Multi-Task Time Series Model Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Timer: Generative Pre-trained Transformers Are Large Time Series Models              Yong Liu, Haoran Zhang, Chenyu Li, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-02-04 ArXiv, DBLP 4 66 visibility_off Large Pre-trained time series models for cross-domain Time series analysis tasks              Harshavardhan Kamarthi, B. A. Prakash            2023-11-19 ArXiv 2 7 visibility_off TOTEM: TOkenized Time Series EMbeddings for General Time Series Analysis              Sabera Talukder, Yisong Yue, Georgia Gkioxari            2024-02-26 ArXiv 4 4 visibility_off Universal Time-Series Representation Learning: A Survey              Patara Trirat, Yooju Shin, Junhyeok Kang, Youngeun Nam, Jihye Na, Minyoung Bae, Joeun Kim, Byunghyun Kim, Jae-Gil Lee            2024-01-08 ArXiv 5 6 visibility_off One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors              Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin            2023-11-24 ArXiv 4 7 visibility_off Chronos: Learning the Language of Time Series              Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang            2024-03-12 ArXiv 28 18 visibility_off Unified Training of Universal Time Series Forecasting Transformers              Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, Doyen Sahoo            2024-02-04 ArXiv 32 22 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"}]}
\ No newline at end of file
+{"config":{"lang":["en"],"separator":"[\\s\\-]+","pipeline":["stopWordFilter"]},"docs":[{"location":"","title":"Home","text":"Introduction <p>Welcome to Team VPE's Literature Survey System! This project leverages the powerful Semantic Scholar's Recommendation API to provide you with highly relevant research article recommendations based on your curated lists of articles.</p> <p>Moreover, this website contains curated collection of references on differentiable and learned simulator algorithms for developing digital twins with applications in drug discovery and development.</p> Features <ul> <li>List-based recommendations: For each topic, we use a manually curated list of positive articles. We then select an equal number of negative articles from other topics. Semantics Scholar's recommendation system uses the set of positive and negative to provide up to 300 recommendations per topic.</li> <li>Single-paper recommendations: For each article in our curated lists, Semantics Scholar's recommendation system returns up to 10 related articles.</li> </ul> How it works? <ul> <li>Scroll to the top of this page and click one of the tabs to open the relevant section.</li> <li>Each topic contains two tables. The first table lists the manually curated articles, and the second table lists the articles recommended for that topic by the recommendation system. The last column of the first table contains a hyperlink that opens a new tab with articles recommended for that specific article.</li> <li>Click on the help icon at the top of the page to learn more about that section.</li> </ul>         Check out our GitHub open_in_new repository to learn how to build one for yourself."},{"location":"Symbolic%20regression/","title":"Symbolic regression","text":"<p> This page was last updated on 2024-11-11 06:06:01 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p> </p> <p> </p>"},{"location":"Symbolic%20regression/#manually_curated_articles","title":"Manually curated articles on Symbolic regression","text":"Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index                View recommendations                visibility_off                  Discovering governing equations from data by sparse identification of nonlinear dynamical systems                               S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences of the United States of America, Proceedings of the National Academy of Sciences 3408 67 open_in_new visibility_off                  Robust learning from noisy, incomplete, high-dimensional experimental data via physically constrained symbolic regression                               Patrick A. K. Reinbold, Logan Kageorge, M. Schatz, R. Grigoriev            2021-02-24 Nature Communications 91 24 open_in_new visibility_off                  Data-driven discovery of coordinates and governing equations                               Kathleen P. Champion, Bethany Lusch, J. Kutz, S. Brunton            2019-03-29 Proceedings of the National Academy of Sciences of the United States of America 648 67 open_in_new visibility_off                  Chaos as an intermittently forced linear system                               S. Brunton, Bingni W. Brunton, J. Proctor, E. Kaiser, J. Kutz            2016-08-18 Nature Communications 468 67 open_in_new visibility_off                  Sparse identification of nonlinear dynamics for model predictive control in the low-data limit                               E. Kaiser, J. Kutz, S. Brunton            2017-11-15 Proceedings of the Royal Society A, Proceedings. Mathematical, Physical, and Engineering Sciences 461 67 open_in_new visibility_off                  Inferring Biological Networks by Sparse Identification of Nonlinear Dynamics                               N. Mangan, S. Brunton, J. Proctor, J. Kutz            2016-05-26 IEEE Transactions on Molecular, Biological and Multi-Scale Communications, IEEE Transactions on Molecular Biological and Multi-Scale Communications 334 67 open_in_new visibility_off                  SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics                               Kadierdan Kaheman, J. Kutz, S. Brunton            2020-04-05 Proceedings of the Royal Society A, Proceedings. Mathematical, Physical, and Engineering Sciences 221 67 open_in_new visibility_off                  Multidimensional Approximation of Nonlinear Dynamical Systems                               Patrick Gel\u00df, Stefan Klus, J. Eisert, Christof Schutte            2018-09-07 Journal of Computational and Nonlinear Dynamics 65 79 open_in_new visibility_off                  Learning Discrepancy Models From Experimental Data                               Kadierdan Kaheman, E. Kaiser, B. Strom, J. Kutz, S. Brunton            2019-09-18 ArXiv, arXiv.org 35 67 open_in_new visibility_off                  Discovery of Physics From Data: Universal Laws and Discrepancies                               Brian M. de Silva, D. Higdon, S. Brunton, J. Kutz            2019-06-19 Frontiers in Artificial Intelligence 77 67 open_in_new visibility_off                  Data-driven discovery of partial differential equations                               S. Rudy, S. Brunton, J. Proctor, J. Kutz            2016-09-21 Science Advances 1235 67 open_in_new visibility_off                  Ensemble-SINDy: Robust sparse model discovery in the low-data, high-noise limit, with active learning and control                               Urban Fasel, J. Kutz, Bingni W. Brunton, S. Brunton            2021-11-22 Proceedings of the Royal Society A, Proceedings. Mathematical, Physical, and Engineering Sciences 181 67 open_in_new visibility_off                  Learning sparse nonlinear dynamics via mixed-integer optimization                               D. Bertsimas, Wes Gurnee            2022-06-01 Nonlinear Dynamics 30 92 open_in_new visibility_off                  A Unified Framework for Sparse Relaxed Regularized Regression: SR3                               P. Zheng, T. Askham, S. Brunton, J. Kutz, A. Aravkin            2018-07-14 IEEE Access 124 67 open_in_new Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index View recommendations"},{"location":"Symbolic%20regression/#recommended_articles","title":"Recommended articles on Symbolic regression","text":"Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index visibility_off                LES-SINDy: Laplace-Enhanced Sparse Identification of Nonlinear Dynamical Systems                             Haoyang Zheng, Guang Lin            2024-11-04 ArXiv 0 1 visibility_off                ADAM-SINDy: An Efficient Optimization Framework for Parameterized Nonlinear Dynamical System Identification                             Siva Viknesh, Younes Tatari, Amirhossein Arzani            2024-10-21 ArXiv 0 24 visibility_off                Deep Generative Modeling for Identification of Noisy, Non-Stationary Dynamical Systems                             Doris Voina, S. Brunton, J. Kutz            2024-10-02 ArXiv 0 67 visibility_off                Weak-form modified sparse identification of nonlinear dynamics                             Cristian L'opez, 'Angel Naranjo, Diego Salazar, Keegan J. Moore            2024-10-23 ArXiv 0 1 visibility_off                Data-driven model discovery and model selection for noisy biological systems                             Xiaojun Wu, MeiLu McDermott, Adam L. Maclean            2024-10-04 bioRxiv 0 19 visibility_off                Physics-informed AI and ML-based sparse system identification algorithm for discovery of PDE's representing nonlinear dynamic systems                             A. Pal, Sutanu Bhowmick, Satish Nagarajaiah            2024-10-13 ArXiv 0 9 visibility_off                Interpretable and Efficient Data-driven Discovery and Control of Distributed Systems                             Florian Wolf, Nicol\u00f2 Botteghi, Urban Fasel, Andrea Manzoni            2024-11-06 ArXiv 0 11 visibility_off                Parsimonious Dynamic Mode Decomposition: A Robust and Automated Approach for Optimally Sparse Mode Selection in Complex Systems                             Arpan Das, Pier Marzocca, O. Levinski            2024-10-22 ArXiv 0 9 visibility_off                Governing equation discovery of a complex system from snapshots                             Qunxi Zhu, Bolin Zhao, Jingdong Zhang, Peiyang Li, Wei Lin            2024-10-22 ArXiv 0 10 visibility_off                Automated Global Analysis of Experimental Dynamics through Low-Dimensional Linear Embeddings                             Samuel A. Moore, Brian P. Mann, Boyuan Chen            2024-11-01 ArXiv 0 2 visibility_off                Discovering Governing Equations of Biological Systems through Representation Learning and Sparse Model Discovery                             Mehrshad Sadria, Vasu Swaroop            2024-09-23 bioRxiv 0 8 visibility_off                Data-driven prediction of large-scale spatiotemporal chaos with distributed low-dimensional models                             Ricardo Constante-Amores, Alec J. Linot, Michael D. Graham            2024-10-02 ArXiv 0 8 visibility_off                Online learning in bifurcating dynamic systems via SINDy and Kalman filtering                             Luca Rosafalco, Paolo Conti, Andrea Manzoni, Stefano Mariani, A. Frangi            2024-11-07 ArXiv 0 8 visibility_off                Data-driven model discovery with Kolmogorov-Arnold networks                             Mohammadamin Moradi, Shirin Panahi, E. Bollt, Ying-Cheng Lai            2024-09-23 ArXiv 2 37 visibility_off                An evolutionary approach for discovering non-Gaussian stochastic dynamical systems based on nonlocal Kramers-Moyal formulas                             Yang Li, Shengyuan Xu, Jinqiao Duan            2024-09-29 ArXiv 0 0 visibility_off                Kernel Operator-Theoretic Bayesian Filter for Nonlinear Dynamical Systems                             Kan Li, Jos'e C. Pr'incipe            2024-10-31 ArXiv 0 2 visibility_off                CGKN: A Deep Learning Framework for Modeling Complex Dynamical Systems and Efficient Data Assimilation                             Chuanqi Chen, Nan Chen, Yinling Zhang, Jin-Long Wu            2024-10-26 ArXiv 0 3 visibility_off                Discovery and inversion of the viscoelastic wave equation in inhomogeneous media                             Su Chen, Yi Ding, Hiroe Miyake, Xiaojun Li            2024-09-27 ArXiv 0 2 visibility_off                Fourier Domain Physics Informed Neural Network                             Jonathan Musgrave, Shu-Wei Huang            2024-09-30 ArXiv 0 1 visibility_off                Reinforcement learning-based estimation for spatio-temporal systems                             Saviz Mowlavi, M. Benosman            2024-09-28 Scientific Reports 0 23 visibility_off                Automated Discovery of Continuous Dynamics from Videos                             Kuang Huang, Dong Heon Cho, Boyuan Chen            2024-10-14 ArXiv 1 0 visibility_off                Stability analysis of chaotic systems in latent spaces                             Elise Ozalp, L. Magri            2024-10-01 ArXiv 0 2 visibility_off                Global dynamical structures from infinitesimal data                             Benjamin W. McInroe, R. Full, D. Koditschek, Yuliy M. Baryshnikov            2024-10-03 ArXiv 0 68 visibility_off                Parametric Taylor series based latent dynamics identification neural networks                             Xinlei Lin, Dunhui Xiao            2024-10-05 ArXiv 0 0 visibility_off                Response Estimation and System Identification of Dynamical Systems via Physics-Informed Neural Networks                             M. Haywood-Alexander, Giacamo Arcieri, A. Kamariotis, Eleni Chatzi            2024-10-02 ArXiv 0 4 visibility_off                Reconstructing dynamics from sparse observations with no training on target system                             Zheng-Meng Zhai, Jun-Yin Huang, Benjamin D. Stern, Ying-Cheng Lai            2024-10-28 ArXiv 0 5 visibility_off                Measure-Theoretic Time-Delay Embedding                             Jonah Botvinick-Greenhouse, Maria Oprea, R. Maulik, Yunan Yang            2024-09-13 ArXiv 0 22 visibility_off                Machine learning sparse reaction-diffusion models from stochastic dynamics and spatiotemporal patterns                             Bedri Abubaker-Sharif, Peter N. Devreotes, Pablo A. Iglesias            2024-10-03 bioRxiv 0 4 visibility_off                Modeling chaotic Lorenz ODE System using Scientific Machine Learning                             Sameera S Kashyap, R. Dandekar, R. Dandekar, S. Panat            2024-10-09 ArXiv 0 3 visibility_off                Data-driven discovery of chemotactic migration of bacteria via coordinate-invariant machine learning                             Y. M. Psarellis, Seungjoon Lee, Tapomoy Bhattacharjee, S. Datta, J. M. Bello-Rivas, Ioannis G Kevrekidis            2024-10-24 BMC Bioinformatics 0 35 visibility_off                Physics-informed Kolmogorov-Arnold Network with Chebyshev Polynomials for Fluid Mechanics                             Chunyu Guo, Lucheng Sun, Shilong Li, Zelong Yuan, Chao Wang            2024-11-07 ArXiv 0 1 visibility_off                Data-Driven Discovery of Conservation Laws from Trajectories via Neural Deflation                             Shaoxuan Chen, Panayotis Kevrekidis, Hong-Kun Zhang, Wei Zhu            2024-10-07 ArXiv 0 1 visibility_off                Estimate of Koopman modes and eigenvalues with Kalman Filter                             Ningxin Liu, Shuigen Liu, Xin T. Tong, Lijian Jiang            2024-09-24 ArXiv 0 2 visibility_off                Causal Discovery in Nonlinear Dynamical Systems using Koopman Operators                             Adam Rupe, Derek DeSantis, Craig Bakker, Parvathi Kooloth, Jian Lu            2024-10-14 ArXiv 0 2 visibility_off                Physics-aligned Schr\u00f6dinger bridge                             Zeyu Li, Hongkun Dou, Shen Fang, Wang Han, Yue Deng, Lijun Yang            2024-09-26 ArXiv 0 2 visibility_off                A method for identifying causality in the response of nonlinear dynamical systems                             Joseph Massingham, Ole Nielsen, Tore Butlin            2024-09-26 ArXiv 0 1 visibility_off                Almost-Linear RNNs Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction                             Manuel Brenner, Christoph Jurgen Hemmer, Z. Monfared, Daniel Durstewitz            2024-10-18 ArXiv 0 7 visibility_off                Learning Generalized Hamiltonians using fully Symplectic Mappings                             Harsh Choudhary, Chandan Gupta, Vyacheslav kungrutsev, Melvin Leok, Georgios Korpas            2024-09-17 ArXiv 0 0 visibility_off                Learning Chaotic Dynamics with Embedded Dissipativity                             Sunbochen Tang, T. Sapsis, Navid Azizan            2024-10-01 ArXiv 0 39 visibility_off                Informative Input Design for Dynamic Mode Decomposition                             Joshua Ott, Mykel J. Kochenderfer, Stephen Boyd            2024-09-19 ArXiv 0 17 visibility_off                Barycentric rational approximation for learning the index of a dynamical system from limited data                             Davide Pradovera, I. V. Gosea, Jan Heiland            2024-10-02 ArXiv 0 12 visibility_off                Gaussian Processes simplify differential equations                             Jonghyeon Lee, B. Hamzi, Y. Kevrekidis, H. Owhadi            2024-10-03 ArXiv 0 36 visibility_off                Bridging the Gap between Koopmanism and Response Theory: Using Natural Variability to Predict Forced Response                             Niccol\u00f2 Zagli, Matthew Colbrook, Valerio Lucarini, Igor Mezi'c, John Moroney            2024-10-02 ArXiv 0 5 visibility_off                Efficient pseudometrics for data-driven comparisons of nonlinear dynamical systems                             Bryan Glaz            2024-09-27 ArXiv 0 0 visibility_off                Koopman Spectral Analysis from Noisy Measurements based on Bayesian Learning and Kalman Smoothing                             Zhexuan Zeng, Jun Zhou, Yasen Wang, Zuowei Ping            2024-10-01 ArXiv 0 0 visibility_off                Physics-Informed Echo State Networks for Modeling Controllable Dynamical Systems                             Eric Mochiutti, Eric A. Antonelo, Eduardo Camponogara            2024-09-27 ArXiv 0 14 visibility_off                Discovery of Quasi-Integrable Equations from traveling-wave data using the Physics-Informed Neural Networks                             A. Nakamula, N. Sawado, K. Shimasaki, Y. Shimazaki, Y. Suzuki, K. Toda            2024-10-23 ArXiv 0 1 visibility_off                Poisson-Dirac Neural Networks for Modeling Coupled Dynamical Systems across Domains                             Razmik Arman Khosrovian, Takaharu Yaguchi, Hiroaki Yoshimura, Takashi Matsubara            2024-10-15 ArXiv 0 11 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index"},{"location":"Time-series%20forecasting/","title":"Time-series forecasting","text":"<p> This page was last updated on 2024-11-11 06:05:40 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p> </p> <p> </p>"},{"location":"Time-series%20forecasting/#manually_curated_articles","title":"Manually curated articles on Time-series forecasting","text":"Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index                View recommendations                visibility_off                  A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection                               Ming Jin, Huan Yee Koh, Qingsong Wen, Daniele Zambon, C. Alippi, G. I. Webb, Irwin King, Shirui Pan            2023-07-07 IEEE Transactions on Pattern Analysis and Machine Intelligence 82 50 open_in_new visibility_off                  Graph-Guided Network for Irregularly Sampled Multivariate Time Series                               Xiang Zhang, M. Zeman, Theodoros Tsiligkaridis, M. Zitnik            2021-10-11 ArXiv, International Conference on Learning Representations 86 47 open_in_new visibility_off                  Taming Local Effects in Graph-based Spatiotemporal Forecasting                               Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-02-08 ArXiv, Neural Information Processing Systems 22 50 open_in_new visibility_off                  Sparse Graph Learning from Spatiotemporal Time Series                               Andrea Cini, Daniele Zambon, C. Alippi            2022-05-26 J. Mach. Learn. Res., Journal of machine learning research 14 50 open_in_new visibility_off                  Graph Deep Learning for Time Series Forecasting                               Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-10-24 ArXiv, arXiv.org 9 50 open_in_new visibility_off                  Large Language Models Are Zero-Shot Time Series Forecasters                               Nate Gruver, Marc Finzi, Shikai Qiu, Andrew Gordon Wilson            2023-10-11 ArXiv, Neural Information Processing Systems 198 14 open_in_new visibility_off                  Graph-Mamba: Towards Long-Range Graph Sequence Modeling with Selective State Spaces                               Chloe X. Wang, Oleksii Tsepa, Jun Ma, Bo Wang            2024-02-01 ArXiv, arXiv.org 56 5 open_in_new visibility_off                  A decoder-only foundation model for time-series forecasting                               Abhimanyu Das, Weihao Kong, Rajat Sen, Yichen Zhou            2023-10-14 ArXiv, International Conference on Machine Learning 103 14 open_in_new visibility_off                  Unified Training of Universal Time Series Forecasting Transformers                               Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, Doyen Sahoo            2024-02-04 ArXiv, International Conference on Machine Learning 71 25 open_in_new visibility_off                  Time-LLM: Time Series Forecasting by Reprogramming Large Language Models                               Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen            2023-10-03 ArXiv, International Conference on Learning Representations 195 9 open_in_new visibility_off                  Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series                               Vijay Ekambaram, Arindam Jati, Nam H. Nguyen, Pankaj Dayama, Chandra Reddy, Wesley M. Gifford, Jayant Kalagnanam            2024-01-08 ArXiv, arXiv.org 3 3 open_in_new visibility_off                  Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistency                               Xiang Zhang, Ziyuan Zhao, Theodoros Tsiligkaridis, M. Zitnik            2022-06-17 ArXiv, Neural Information Processing Systems 211 47 open_in_new visibility_off                  Domain Adaptation for Time Series Under Feature and Label Shifts                               Huan He, Owen Queen, Teddy Koker, Consuelo Cuevas, Theodoros Tsiligkaridis, M. Zitnik            2023-02-06 ArXiv, DBLP 38 47 open_in_new visibility_off                  AZ-whiteness test: a test for signal uncorrelation on spatio-temporal graphs                               Daniele Zambon, C. Alippi            None DBLP 6 50 open_in_new visibility_off                  Graph state-space models                               Daniele Zambon, Andrea Cini, L. Livi, C. Alippi            2023-01-04 ArXiv, arXiv.org 4 50 open_in_new visibility_off                  UNITS: A Unified Multi-Task Time Series Model                               Shanghua Gao, Teddy Koker, Owen Queen, Thomas Hartvigsen, Theodoros Tsiligkaridis, M. Zitnik            2024-02-29 ArXiv 4 47 open_in_new Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index View recommendations"},{"location":"Time-series%20forecasting/#recommended_articles","title":"Recommended articles on Time-series forecasting","text":"Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index visibility_off                A Mamba Foundation Model for Time Series Forecasting                             Haoyu Ma, Yushu Chen, Wenlai Zhao, Jinzhe Yang, Yingsheng Ji, Xinghua Xu, Xiaozhu Liu, Hao Jing, Shengzhuo Liu, Guangwen Yang            2024-11-05 ArXiv 0 7 visibility_off                Metadata Matters for Time Series: Informative Forecasting with Transformers                             Jiaxiang Dong, Haixu Wu, Yuxuan Wang, Li Zhang, Jianmin Wang, Mingsheng Long            2024-10-04 ArXiv 0 67 visibility_off                FlexTSF: A Universal Forecasting Model for Time Series with Variable Regularities                             Jingge Xiao, Yile Chen, Gao Cong, Wolfgang Nejdl, Simon Gottschalk            2024-10-30 ArXiv 0 6 visibility_off                Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts                             X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin            2024-09-24 ArXiv 2 9 visibility_off                Mamba4Cast: Efficient Zero-Shot Time Series Forecasting with State Space Models                             Sathya Kamesh Bhethanabhotla, Omar Swelam, Julien N. Siems, David Salinas, Frank Hutter            2024-10-12 ArXiv 0 8 visibility_off                Timer-XL: Long-Context Transformers for Unified Time Series Forecasting                             Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-10-07 ArXiv 0 67 visibility_off                xLSTM-Mixer: Multivariate Time Series Forecasting by Mixing via Scalar Memories                             Maurice Kraus, Felix Divo, D. Dhami, K. Kersting            2024-10-22 ArXiv 0 12 visibility_off                In-Context Fine-Tuning for Time-Series Foundation Models                             Abhimanyu Das, Matthew Faw, Rajat Sen, Yichen Zhou            2024-10-31 ArXiv 0 14 visibility_off                Towards Long-Context Time Series Foundation Models                             Nina Zukowska, Mononito Goswami, Micha\u0142 Wili\u0144ski, Willa Potosnak, Artur Dubrawski            2024-09-20 ArXiv 0 8 visibility_off                Moirai-MoE: Empowering Time Series Foundation Models with Sparse Mixture of Experts                             Xu Liu, Juncheng Liu, Gerald Woo, Taha \u0130brahim Aksu, Yuxuan Liang, Roger Zimmermann, Chenghao Liu, Silvio Savarese, Caiming Xiong, Doyen Sahoo            2024-10-14 ArXiv 0 25 visibility_off                LLM-Mixer: Multiscale Mixing in LLMs for Time Series Forecasting                             Md. Kowsher, Md. Shohanur, Islam Sobuj, Nusrat Jahan Prottasha, E. A. Alanis, O. Garibay, Niloofar Yousefi, Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timoth\u00e9e Lacroix, Baptiste Rozi\u00e8re, Naman Goyal, Eric Hambro, Xuan Zhang, Xun Liang, Aakas Zhiyuli, Shusen Zhang, Rui Xu, Bo Wu. 2019, Huaqin Zhao, Zheng Liu, Zihao Wu, Yiwei Li, Tianze Yang, Peng Shu, Shaochen Xu, Haixing Dai, Lin Zhao, Gengchen Mai, Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, Wan Zhang            2024-10-15 ArXiv 0 34 visibility_off                Diffusion Auto-regressive Transformer for Effective Self-supervised Time Series Forecasting                             Daoyu Wang, Mingyue Cheng, Zhiding Liu, Qi Liu, Enhong Chen            2024-10-08 ArXiv 1 11 visibility_off                On the Regularization of Learnable Embeddings for Time Series Processing                             L. Butera, G. Felice, Andrea Cini, C. Alippi            2024-10-18 ArXiv 0 50 visibility_off                PSformer: Parameter-efficient Transformer with Segment Attention for Time Series Forecasting                             Yanlong Wang, Jian Xu, Fei Ma, Shao-Lun Huang, Danny D. Sun, Xiao-Ping Zhang            2024-11-03 ArXiv 0 1 visibility_off                Revisited Large Language Model for Time Series Analysis through Modality Alignment                             L. Zheng, C. Dong, W. Zhang, Lin Yue, Miao Xu, Olaf Maennel, Weitong Chen            2024-10-16 ArXiv 0 7 visibility_off                Test Time Learning for Time Series Forecasting                             Panayiotis Christou, Shichu Chen, Xupeng Chen, Parijat Dube            2024-09-21 ArXiv 0 4 visibility_off                Continuous-Time Linear Positional Embedding for Irregular Time Series Forecasting                             Byunghyun Kim, Jae-Gil Lee            2024-09-30 ArXiv 0 2 visibility_off                Approximate attention with MLP: a pruning strategy for attention-based model in multivariate time series forecasting                             Suhan Guo, Jiahong Deng, Yi Wei, Hui Dou, F. Shen, Jian Zhao            2024-10-31 ArXiv 0 7 visibility_off                TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis                             Shiyu Wang, Jiawei Li, X. Shi, Zhou Ye, Baichuan Mo, Wenze Lin, Shengtong Ju, Zhixuan Chu, Ming Jin            2024-10-21 ArXiv 0 3 visibility_off                Towards Generalisable Time Series Understanding Across Domains                             Ozgun Turgut, Philip Muller, M. Menten, Daniel Rueckert            2024-10-09 ArXiv 0 15 visibility_off                Implicit Reasoning in Deep Time Series Forecasting                             Willa Potosnak, Cristian Challu, Mononito Goswami, Micha\u0142 Wili\u0144ski, Nina Zukowska            2024-09-17 ArXiv 0 8 visibility_off                Learning Pattern-Specific Experts for Time Series Forecasting Under Patch-level Distribution Shift                             Yanru Sun, Zongxia Xie, Emadeldeen Eldele, Dongyue Chen, Qinghua Hu, Min Wu            2024-10-13 ArXiv 0 18 visibility_off                FilterNet: Harnessing Frequency Filters for Time Series Forecasting                             Kun Yi, Jingru Fei, Qi Zhang, Hui He, Shufeng Hao, Defu Lian, Wei Fan            2024-11-03 ArXiv 0 6 visibility_off                AutoSTF: Decoupled Neural Architecture Search for Cost-Effective Automated Spatio-Temporal Forecasting                             Tengfei Lyu, Weijiao Zhang, Jinliang Deng, Hao Liu            2024-09-25 ArXiv 0 7 visibility_off                ElasTST: Towards Robust Varied-Horizon Forecasting with Elastic Time-Series Transformer                             Jiawen Zhang, Shun Zheng, Xumeng Wen, Xiaofang Zhou, Jiang Bian, Jia Li            2024-11-04 ArXiv 0 8 visibility_off                Self-Supervised Learning of Disentangled Representations for Multivariate Time-Series                             Ching Chang, Chiao-Tung Chan, Wei-Yao Wang, Wenjie Peng, Tien-Fu Chen            2024-10-16 ArXiv 0 7 visibility_off                CIPformer: Channel-Independence Integrated with Patching Transformer for Multivariate Long-Term Time Series Forecasting                             Peng Qiu, Jinhong Li            2024-09-20 2024 IEEE 7th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) 0 0 visibility_off                UmambaTSF: A U-shaped Multi-Scale Long-Term Time Series Forecasting Method Using Mamba                             Li Wu, Wenbin Pei, Jiulong Jiao, Qiang Zhang            2024-10-15 ArXiv 0 0 visibility_off                ExoTST: Exogenous-Aware Temporal Sequence Transformer for Time Series Prediction                             Kshitij Tayal, Arvind Renganathan, X. Jia, Vipin Kumar, Dan Lu            2024-10-16 ArXiv 0 19 visibility_off                Integration of Mamba and Transformer - MAT for Long-Short Range Time Series Forecasting with Application to Weather Dynamics                             Wenqing Zhang, Junming Huang, Ruotong Wang, Changsong Wei, Wenqian Huang,             2024-09-13 ArXiv 0 2 visibility_off                Towards Neural Scaling Laws for Time Series Foundation Models                             Qingren Yao, Chao-Han Huck Yang, Renhe Jiang, Yuxuan Liang, Ming Jin, Shirui Pan            2024-10-16 ArXiv 0 6 visibility_off                TVGCN: Time-varying graph convolutional networks for multivariate and multifeature spatiotemporal series prediction                             Feiyan Sun, Wenning Hao, Ao Zou, Kai Cheng            2024-07-01 Science Progress 0 2 visibility_off                TiVaT: Joint-Axis Attention for Time Series Forecasting with Lead-Lag Dynamics                             Junwoo Ha, Hyukjae Kwon, Sungsoo Kim, Kisu Lee, Ha Young Kim            2024-10-02 ArXiv 0 3 visibility_off                Sequential Order-Robust Mamba for Time Series Forecasting                             Seunghan Lee, Juri Hong, Kibok Lee, Taeyoung Park            2024-10-30 ArXiv 0 2 visibility_off                DisenTS: Disentangled Channel Evolving Pattern Modeling for Multivariate Time Series Forecasting                             Zhiding Liu, Jiqian Yang, Qingyang Mao, Yuze Zhao, Mingyue Cheng, Zhi Li, Qi Liu, Enhong Chen            2024-10-30 ArXiv 0 5 visibility_off                Irregularity-Informed Time Series Analysis: Adaptive Modelling of Spatial and Temporal Dynamics                             L. Zheng, Zhengyang Li, C. Dong, W. Zhang, Lin Yue, Miao Xu, Olaf Maennel, Weitong Chen            2024-10-16 Proceedings of the 33rd ACM International Conference on Information and Knowledge Management 0 7 visibility_off                TS-TCD: Triplet-Level Cross-Modal Distillation for Time-Series Forecasting Using Large Language Models                             Pengfei Wang, Huanran Zheng, Silong Dai, Wenjing Yue, Wei Zhu, Xiaoling Wang            2024-09-23 ArXiv 1 3 visibility_off                Autoregressive Moving-average Attention Mechanism for Time Series Forecasting                             Jiecheng Lu, Xu Han, Yan Sun, Shihao Yang            2024-10-04 ArXiv 0 2 visibility_off                D2Vformer: A Flexible Time Series Prediction Model Based on Time Position Embedding                             Xiaobao Song, Hao Wang, Liwei Deng, Yuxin He, Wenming Cao, Andrew Chi-Sing Leung            2024-09-17 ArXiv 0 1 visibility_off                TrajGPT: Irregular Time-Series Representation Learning for Health Trajectory Analysis                             Ziyang Song, Qingcheng Lu, He Zhu, David L. Buckeridge, Yue Li            2024-10-03 ArXiv 0 2 visibility_off                TimeCNN: Refining Cross-Variable Interaction on Time Point for Time Series Forecasting                             Ao Hu, Dongkai Wang, Yong Dai, Shiyi Qi, Liangjian Wen, Jun Wang, Zhi Chen, Xun Zhou, Zenglin Xu, Jiang Duan            2024-10-07 ArXiv 0 3 visibility_off                MMFNet: Multi-Scale Frequency Masking Neural Network for Multivariate Time Series Forecasting                             Aitian Ma, Dongsheng Luo, Mo Sha            2024-10-02 ArXiv 0 2 visibility_off                LLM-TS Integrator: Integrating LLM for Enhanced Time Series Modeling                             Can Chen, Gabriel Oliveira, H. S. Noghabi, Tristan Sylvain            2024-10-21 ArXiv 0 2 visibility_off                A Temporal Linear Network for Time Series Forecasting                             Remi Genet, Hugo Inzirillo            2024-10-28 ArXiv 0 2 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index"},{"location":"recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378/","title":"06a0ba437d41a7c82c08a9636a4438c1b5031378","text":""},{"location":"recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378/#_1","title":"06a0ba437d41a7c82c08a9636a4438c1b5031378","text":"<p> This page was last updated on 2024-11-11 06:05:45 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Inferring Biological Networks by Sparse Identification of Nonlinear Dynamics Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Distilling identifiable and interpretable dynamic models from biological data              Gemma Massonis, A. F. Villaverde, J. Banga            2023-03-15 PLOS Computational Biology 5 55 visibility_off Characterization of Biologically Relevant Network Structures form Time-series Data              Z. Tuza, G. Stan            2018-09-24 2018 IEEE Conference on Decision and Control (CDC) 3 35 visibility_off Reconstruction of arbitrary biochemical reaction networks: A compressive sensing approach              W. Pan, Ye Yuan, G. Stan            2012-05-08 2012 IEEE 51st IEEE Conference on Decision and Control (CDC) 33 35 visibility_off Inferring sparse networks for noisy transient processes              H. M. Tran, S. Bukkapatnam            2016-02-26 Scientific Reports 6 34 visibility_off Koopman Operators for Generalized Persistence of Excitation Conditions for Nonlinear Systems              N. Boddupalli, A. Hasnain, S. Nandanoori, Enoch Yeung            2019-06-25 2019 IEEE 58th Conference on Decision and Control (CDC) 5 17 visibility_off Learning sparse nonlinear dynamics via mixed-integer optimization              D. Bertsimas, Wes Gurnee            2022-06-01 Nonlinear Dynamics 30 92 visibility_off Inverse problems in systems biology              H. Engl, Christoph Flamm, P. K\u00fcgler, James Lu, S. M\u00fcller, P. Schuster            2009-12-01 Inverse Problems 136 48 visibility_off Data-driven model discovery and model selection for noisy biological systems              Xiaojun Wu, MeiLu McDermott, Adam L. Maclean            2024-10-04 bioRxiv 0 19 visibility_off Set-base dynamical parameter estimation and model invalidation for biochemical reaction networks              Philipp Rumschinski, S. Borchers, S. Bosio, R. Weismantel, R. Findeisen            2010-05-25 BMC Systems Biology 67 48 visibility_off Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems              Mohammad Amin Basiri, Sina Khanmohammadi            2024-09-02 ArXiv 0 3 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/0acd117521ef5aafb09fed02ab415523b330b058/","title":"0acd117521ef5aafb09fed02ab415523b330b058","text":""},{"location":"recommendations/0acd117521ef5aafb09fed02ab415523b330b058/#_1","title":"0acd117521ef5aafb09fed02ab415523b330b058","text":"<p> This page was last updated on 2024-11-11 06:05:51 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Data-driven discovery of partial differential equations Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Machine discovery of partial differential equations from spatiotemporal data: A sparse Bayesian learning framework.              Ye Yuan, Xiuting Li, Liang Li, Frank J. Jiang, Xiuchuan Tang, Fumin Zhang, Jorge Gon\u00e7alves, Henning U. Voss, Han Ding, J\u00fcrgen Kurths            2023-11-01 Chaos 4 11 visibility_off Machine Discovery of Partial Differential Equations from Spatiotemporal Data              Ye Yuan, Junlin Li, Liang Li, Frank Jiang, Xiuchuan Tang, Fumin Zhang, Sheng Liu, J. Gon\u00e7alves, H. Voss, Xiuting Li, J. Kurths, Han Ding            2019-09-15 ArXiv 9 107 visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3408 67 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 11 2 visibility_off Physics-informed learning of governing equations from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 Nature Communications 289 13 visibility_off Sparsistent Model Discovery              Georges Tod, G. Both, R. Kusters            2021-06-22 ArXiv 1 13 visibility_off Supplementary material from \"Learning partial differential equations via data discovery and sparse optimization\"              Hayden Schaeffer            2017-01-16 0 16 visibility_off Data-Driven discovery of governing physical laws and their parametric dependencies in engineering, physics and biology              J. Kutz, Samuel H. Rudy, A. Alla, S. Brunton            2017-12-01 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP) 12 67 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b/","title":"0d01d21137a5af9f04e4b16a55a0f732cb8a540b","text":""},{"location":"recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b/#_1","title":"0d01d21137a5af9f04e4b16a55a0f732cb8a540b","text":"<p> This page was last updated on 2024-11-11 06:05:10 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Sparse Graph Learning from Spatiotemporal Time Series Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Graph Deep Learning for Time Series Forecasting              Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-10-24 ArXiv 9 50 visibility_off Sparsity exploitation via discovering graphical models in multi-variate time-series forecasting              Ngoc-Dung Do, T. Hy, D. Nguyen            2023-06-29 ArXiv 0 7 visibility_off TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting              Nancy R. Xu, Chrysoula Kosma, M. Vazirgiannis            2023-07-27 ArXiv 1 55 visibility_off Balanced Graph Structure Learning for Multivariate Time Series Forecasting              Weijun Chen, Yanze Wang, Chengshuo Du, Zhenglong Jia, Feng Liu, Ran Chen            2022-01-24 ArXiv 0 11 visibility_off A Study of Joint Graph Inference and Forecasting              Daniel Z\u00fcgner, Fran\u00e7ois-Xavier Aubet, Victor Garcia Satorras, Tim Januschowski, Stephan G\u00fcnnemann, Jan Gasthaus            2021-09-10 ArXiv 12 47 visibility_off Regularized Graph Structure Learning with Semantic Knowledge for Multi-variates Time-Series Forecasting              Hongyuan Yu, Ting Li, Weichen Yu, Jianguo Li, Yan Huang, Liang Wang, A. Liu            2022-07-01 ArXiv, DBLP 44 35 visibility_off Discrete Graph Structure Learning for Forecasting Multiple Time Series              Chao Shang, Jie Chen, J. Bi            2021-01-18 ArXiv 199 38 visibility_off Learning Graph Structures and Uncertainty for Accurate and Calibrated Time-series Forecasting              Harshavardhan Kamarthi, Lingkai Kong, Alexander Rodr\u00edguez, Chao Zhang, B. A. Prakash            2024-07-02 ArXiv 0 9 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/123acfbccca0460171b6b06a4012dbb991cde55b/","title":"123acfbccca0460171b6b06a4012dbb991cde55b","text":""},{"location":"recommendations/123acfbccca0460171b6b06a4012dbb991cde55b/#_1","title":"123acfbccca0460171b6b06a4012dbb991cde55b","text":"<p> This page was last updated on 2024-11-11 06:05:12 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Large Language Models Are Zero-Shot Time Series Forecasters Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Chronos: Learning the Language of Time Series              Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang            2024-03-12 ArXiv 69 18 visibility_off AutoTimes: Autoregressive Time Series Forecasters via Large Language Models              Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-02-04 ArXiv 5 67 visibility_off An Evaluation of Standard Statistical Models and LLMs on Time Series Forecasting              Rui Cao, Qiao Wang            2024-08-09 ArXiv 0 3 visibility_off Are Language Models Actually Useful for Time Series Forecasting?              Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen            2024-06-22 ArXiv 13 3 visibility_off Timer: Generative Pre-trained Transformers Are Large Time Series Models              Yong Liu, Haoran Zhang, Chenyu Li, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-02-04 ArXiv, DBLP 21 67 visibility_off TimeGPT-1              Azul Garza, Cristian Challu, Max Mergenthaler-Canseco            2023-10-05 ArXiv 60 5 visibility_off MultiCast: Zero-Shot Multivariate Time Series Forecasting Using LLMs              Georgios Chatzigeorgakidis, Konstantinos Lentzos, Dimitrios Skoutas            2024-05-13 2024 IEEE 40th International Conference on Data Engineering Workshops (ICDEW) 0 7 visibility_off Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilities              Mingyu Jin, Hua Tang, Chong Zhang, Qinkai Yu, Chengzhi Liu, Suiyuan Zhu, Yongfeng Zhang, Mengnan Du            2024-02-16 ArXiv 5 8 visibility_off Time-LLM: Time Series Forecasting by Reprogramming Large Language Models              Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen            2023-10-03 ArXiv 195 9 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277/","title":"16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277","text":""},{"location":"recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277/#_1","title":"16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277","text":"<p> This page was last updated on 2024-11-11 06:05:18 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Time-LLM: Time Series Forecasting by Reprogramming Large Language Models Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Time-FFM: Towards LM-Empowered Federated Foundation Model for Time Series Forecasting              Qingxiang Liu, Xu Liu, Chenghao Liu, Qingsong Wen, Yuxuan Liang            2024-05-23 ArXiv 0 7 visibility_off Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilities              Mingyu Jin, Hua Tang, Chong Zhang, Qinkai Yu, Chengzhi Liu, Suiyuan Zhu, Yongfeng Zhang, Mengnan Du            2024-02-16 ArXiv 5 8 visibility_off LSTPrompt: Large Language Models as Zero-Shot Time Series Forecasters by Long-Short-Term Prompting              Haoxin Liu, Zhiyuan Zhao, Jindong Wang, Harshavardhan Kamarthi, B. A. Prakash            2024-02-25 ArXiv 11 8 visibility_off PromptCast: A New Prompt-Based Learning Paradigm for Time Series Forecasting              Hao Xue, Flora D.Salim            2022-09-20 IEEE Transactions on Knowledge and Data Engineering 80 17 visibility_off Understanding Different Design Choices in Training Large Time Series Models              Yu-Neng Chuang, Songchen Li, Jiayi Yuan, Guanchu Wang, Kwei-Herng Lai, Leisheng Yu, Sirui Ding, Chia-yuan Chang, Qiaoyu Tan, D. Zha, Xia Hu            2024-06-20 ArXiv 1 24 visibility_off A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model              Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung            2024-05-03 ArXiv 7 47 visibility_off In-context Time Series Predictor              Jiecheng Lu, Yan Sun, Shihao Yang            2024-05-23 ArXiv 2 2 visibility_off LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters              Ching Chang, Wenjie Peng, Tien-Fu Chen            2023-08-16 ArXiv 22 3 visibility_off Towards Time Series Reasoning with LLMs              Winnie Chow, Lauren Gardiner, Haraldur T. Hallgr'imsson, Maxwell A. Xu, Shirley You Ren            2024-09-17 ArXiv 2 1 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f/","title":"1df04f33a8ef313cc2067147dbb79c3ca7c5c99f","text":""},{"location":"recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f/#_1","title":"1df04f33a8ef313cc2067147dbb79c3ca7c5c99f","text":"<p> This page was last updated on 2024-11-11 06:05:13 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Graph-Mamba: Towards Long-Range Graph Sequence Modeling with Selective State Spaces Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off HeteGraph-Mamba: Heterogeneous Graph Learning via Selective State Space Model              Zhenyu Pan, Yoonsung Jeong, Xiaoda Liu, Han Liu            2024-05-22 ArXiv 1 6 visibility_off Graph Mamba: Towards Learning on Graphs with State Space Models              Ali Behrouz, Farnoosh Hashemi            2024-02-13 ArXiv, DBLP 39 9 visibility_off Learning Long Range Dependencies on Graphs via Random Walks              Dexiong Chen, Till Hendrik Schulz, Karsten Borgwardt            2024-06-05 ArXiv 2 8 visibility_off What Can We Learn from State Space Models for Machine Learning on Graphs?              Yinan Huang, Siqi Miao, Pan Li            2024-06-09 ArXiv 1 3 visibility_off Context Sketching for Memory-efficient Graph Representation Learning              Kai-Lang Yao, Wusuo Li            2023-12-01 2023 IEEE International Conference on Data Mining (ICDM) 0 4 visibility_off A Scalable and Effective Alternative to Graph Transformers              Kaan Sancak, Zhigang Hua, Jin Fang, Yan Xie, Andrey Malevich, Bo Long, M. F. Balin, \u00dcmit V. \u00c7ataly\u00fcrek            2024-06-17 ArXiv 0 7 visibility_off Hierarchical Graph Transformer with Adaptive Node Sampling              Zaixin Zhang, Qi Liu, Qingyong Hu, Cheekong Lee            2022-10-08 ArXiv 63 19 visibility_off Deformable Graph Transformer              Jinyoung Park, Seongjun Yun, Hyeon-ju Park, Jaewoo Kang, Jisu Jeong, KyungHyun Kim, Jung-Woo Ha, Hyunwoo J. Kim            2022-06-29 ArXiv 6 28 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68/","title":"279cd637b7e38bba1dd8915b5ce68cbcacecbe68","text":""},{"location":"recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68/#_1","title":"279cd637b7e38bba1dd8915b5ce68cbcacecbe68","text":"<p> This page was last updated on 2024-11-11 06:05:28 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Graph state-space models Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Probabilistic Recurrent State-Space Models              Andreas Doerr, Christian Daniel, Martin Schiegg, D. Nguyen-Tuong, S. Schaal, Marc Toussaint, Sebastian Trimpe            2018-01-31 ArXiv, MAG, DBLP 113 93 visibility_off Interpretable Latent Variables in Deep State Space Models              Haoxuan Wu, David S. Matteson, M. Wells            2022-03-03 ArXiv 0 40 visibility_off FACTS: A Factored State-Space Framework For World Modelling              Li Nanbo, Firas Laakom, Yucheng Xu, Wenyi Wang, Jurgen Schmidhuber            2024-10-28 ArXiv 0 7 visibility_off Stanza: A Nonlinear State Space Model for Probabilistic Inference in Non-Stationary Time Series              Anna K. Yanchenko, S. Mukherjee            2020-06-11 ArXiv 6 32 visibility_off Sparse Graph Learning from Spatiotemporal Time Series              Andrea Cini, Daniele Zambon, C. Alippi            2022-05-26 J. Mach. Learn. Res. 14 50 visibility_off Multi-Knowledge Fusion Network for Time Series Representation Learning              Sakhinana Sagar Srinivas, Shivam Gupta, Krishna Sai Sudhir Aripirala, Venkataramana Runkana            2024-08-22 ArXiv 0 3 visibility_off Relational State-Space Model for Stochastic Multi-Object Systems              Fan Yang, Ling Chen, Fan Zhou, Yusong Gao, Wei Cao            2020-01-13 ArXiv 8 60 visibility_off State Space Models on Temporal Graphs: A First-Principles Study              Jintang Li, Ruofan Wu, Xinzhou Jin, Boqun Ma, Liang Chen, Zibin Zheng            2024-06-03 ArXiv 1 12 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196/","title":"2b2aa13d4959073f61ad70555bc8c7da7d116196","text":""},{"location":"recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196/#_1","title":"2b2aa13d4959073f61ad70555bc8c7da7d116196","text":"<p> This page was last updated on 2024-11-11 06:05:47 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Multidimensional Approximation of Nonlinear Dynamical Systems Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off A dynamic mode decomposition extension for the forecasting of parametric dynamical systems              Francesco Andreuzzi, N. Demo, G. Rozza            2021-10-18 ArXiv 21 51 visibility_off Dynamic tensor time series modeling and analysis              A. Surana, G. Patterson, I. Rajapakse            2016-12-01 2016 IEEE 55th Conference on Decision and Control (CDC) 8 24 visibility_off Tensor Train Based Higher Order\u00a0Dynamic Mode Decomposition for Dynamical Systems              Keren Li, S. Utyuzhnikov            2023-04-11 SSRN Electronic Journal 3 12 visibility_off Modeling of dynamical systems through deep learning              P. Rajendra, V. Brahmajirao            2020-11-22 Biophysical Reviews 36 5 visibility_off A dynamical systems based framework for dimension reduction              Ryeongkyung Yoon, B. Osting            2022-04-18 ArXiv 1 19 visibility_off Learning Data-Driven Model of Damped Coupled Oscillators from System Impulse Response              Jacob Fabro, G. Vogl, Yongzhi Qu, Reese Eischens            2022-10-28 Annual Conference of the PHM Society 0 19 visibility_off Deep Neural Networks for Nonlinear Model Order Reduction of Unsteady Flows              Hamidreza Eivazi, H. Veisi, M. H. Naderi, V. Esfahanian            2020-07-02 ArXiv 136 23 visibility_off Nonlinear system identification with regularized Tensor Network B-splines              Ridvan Karagoz, Kim Batselier            2020-03-17 Autom. 18 4 visibility_off Data-driven Reconstruction of Nonlinear Dynamics from Sparse Observation              K. Yeo            2019-06-10 J. Comput. Phys. 23 19 visibility_off Data-driven prediction in dynamical systems: recent developments              Amin Ghadami, B. Epureanu            2022-06-20 Philosophical transactions. Series A, Mathematical, physical, and engineering sciences 48 33 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220/","title":"35e2571c17246577e0bc1b9de57a314c3b60e220","text":""},{"location":"recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220/#_1","title":"35e2571c17246577e0bc1b9de57a314c3b60e220","text":"<p> This page was last updated on 2024-11-11 06:05:50 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Discovery of Physics From Data: Universal Laws and Discrepancies Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off A Unified Sparse Optimization Framework to Learn Parsimonious Physics-Informed Models From Data              Kathleen P. Champion, P. Zheng, A. Aravkin, S. Brunton, J. Kutz            2019-06-25 IEEE Access 103 67 visibility_off Machine-Learning Non-Conservative Dynamics for New-Physics Detection              Ziming Liu, Bohan Wang, Qi Meng, Wei Chen, M. Tegmark, Tie-Yan Liu            2021-05-31 Physical review. E 15 83 visibility_off Discovering conservation laws from trajectories via machine learning              Seungwoong Ha, Hawoong Jeong            2021-02-08 ArXiv 10 43 visibility_off The Lie Detector.              A. Young, A. Lawrie            2019-12-13 arXiv: Signal Processing 0 18 visibility_off Physics-Informed Machine Learning for Modeling and Control of Dynamical Systems              Truong X. Nghiem, J\u00e1n Drgo\u0148a, Colin N. Jones, Zolt\u00e1n Nagy, Roland Schwan, Biswadip Dey, A. Chakrabarty, S. D. Cairano, J. Paulson, Andrea Carron, M. Zeilinger, Wenceslao Shaw-Cortez, D. Vrabie            2023-05-31 2023 American Control Conference (ACC) 22 38 visibility_off Scale-invariant Learning by Physics Inversion              Philipp Holl, V. Koltun, Nils Thuerey            2021-09-30 ArXiv, DBLP 5 105 visibility_off Learning continuous models for continuous physics              A. Krishnapriyan, A. Queiruga, N. Benjamin Erichson, Michael W. Mahoney            2022-02-17 Communications Physics 25 31 visibility_off Machine Learning Conservation Laws from Trajectories.              Ziming Liu, Max Tegmark            2021-05-06 Physical review letters 93 82 visibility_off From data to conservation laws              F. Chirigati            2023-09-01 Nature Computational Science 1 17 visibility_off Physics-informed regularization and structure preservation for learning stable reduced models from data with operator inference              N. Sawant, B. Kramer, B. Peherstorfer            2021-07-06 ArXiv 25 28 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/3c9961153493370500020c81527b3548c96f81e0/","title":"3c9961153493370500020c81527b3548c96f81e0","text":""},{"location":"recommendations/3c9961153493370500020c81527b3548c96f81e0/#_1","title":"3c9961153493370500020c81527b3548c96f81e0","text":"<p> This page was last updated on 2024-11-11 06:05:42 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Data-driven discovery of coordinates and governing equations Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3408 67 visibility_off Physics-informed learning of governing equations from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 Nature Communications 289 13 visibility_off Deep learning of physical laws from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 ArXiv 19 13 visibility_off Modeling of dynamical systems through deep learning              P. Rajendra, V. Brahmajirao            2020-11-22 Biophysical Reviews 36 5 visibility_off Bayesian autoencoders for data-driven discovery of coordinates, governing equations and fundamental constants              Liyao (Mars) Gao, J. Kutz            2022-11-19 Proceedings of the Royal Society A 13 32 visibility_off Symbolic regression via neural networks.              N. Boddupalli, T. Matchen, J. Moehlis            2023-08-01 Chaos 4 37 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 11 2 visibility_off Discovering sparse interpretable dynamics from partial observations              Peter Y. Lu, Joan Ari\u00f1o Bernad, M. Solja\u010di\u0107            2021-07-22 Communications Physics 20 95 visibility_off Sparse Estimation for Hamiltonian Mechanics              Yuya Note, Masahito Watanabe, Hiroaki Yoshimura, Takaharu Yaguchi, Toshiaki Omori            2024-03-25 Mathematics 0 11 visibility_off Uncovering Closed-form Governing Equations of Nonlinear Dynamics from Videos              Lele Luan, Yang Liu, Hao Sun            2021-06-09 ArXiv 0 13 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2/","title":"3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2","text":""},{"location":"recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2/#_1","title":"3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2","text":"<p> This page was last updated on 2024-11-11 06:05:42 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Chaos as an intermittently forced linear system Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Featurizing Koopman mode decomposition for robust forecasting              D. Aristoff, J. Copperman, Nathan Mankovich, Alexander Davies            2023-12-14 The Journal of Chemical Physics 0 12 visibility_off Deep learning delay coordinate dynamics for chaotic attractors from partial observable data              Charles D. Young, M. Graham            2022-11-20 Physical review. E 10 51 visibility_off Discovering governing equations from partial measurements with deep delay autoencoders              Joseph Bakarji, Kathleen P. Champion, J. Nathan Kutz, S. Brunton            2022-01-13 Proceedings of the Royal Society A 75 67 visibility_off DEFM: Delay-embedding-based forecast machine for time series forecasting by spatiotemporal information transformation.              Hao Peng, Pei Chen, R. Liu            2020-05-16 Chaos 1 83 visibility_off DEFM: Delay-embedding-based forecast machine for time series forecasting by spatiotemporal information transformation.              Hao Peng, Pei Chen, R. Liu            2020-05-16 Chaos 1 83 visibility_off Detecting chaos in lineage-trees: A deep learning approach              H. Rappeport, Irit Levin Reisman, Naftali Tishby, N. Balaban            2021-06-08 ArXiv 3 56 visibility_off Cluster-based network modeling\u2014From snapshots to complex dynamical systems              Daniel Fernex, B. R. Noack, R. Semaan            2021-06-01 Science Advances 50 50 visibility_off Cluster-based network modeling -- automated robust modeling of complex dynamical systems              Daniel Fernex, B. R. Noack, R. Semaan            2020-10-30 arXiv: Data Analysis, Statistics and Probability 1 50 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba/","title":"455bfc515eb279cc09023faa1f78c6efb61224ba","text":""},{"location":"recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba/#_1","title":"455bfc515eb279cc09023faa1f78c6efb61224ba","text":"<p> This page was last updated on 2024-11-11 06:05:08 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Graph-Guided Network for Irregularly Sampled Multivariate Time Series Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off PrimeNet: Pre-training for Irregular Multivariate Time Series              Ranak Roy Chowdhury, Jiacheng Li, Xiyuan Zhang, Dezhi Hong, Rajesh K. Gupta, Jingbo Shang            2023-06-26 DBLP 12 17 visibility_off Finding Short Signals in Long Irregular Time Series with Continuous-Time Attention Policy Networks              Thomas Hartvigsen, Jidapa Thadajarassiri, Xiangnan Kong, Elke A. Rundensteiner            2023-02-08 ArXiv 1 39 visibility_off Continuous Time Evidential Distributions for Irregular Time Series              Taylor W. Killian, Haoran Zhang, Thomas Hartvigsen, Ava P. Amini            2023-07-25 ArXiv 0 12 visibility_off Time Series as Images: Vision Transformer for Irregularly Sampled Time Series              Zekun Li, SHIYANG LI, Xifeng Yan            2023-03-01 ArXiv 17 65 visibility_off Time Series as Images: Vision Transformer for Irregularly Sampled Time Series              Zekun Li, SHIYANG LI, Xifeng Yan            2023-03-01 ArXiv 17 65 visibility_off EMIT- Event-Based Masked Auto Encoding for Irregular Time Series              Hrishikesh Patel, Ruihong Qiu, Adam Irwin, S. Sadiq, Sen Wang            2024-09-25 ArXiv 0 2 visibility_off Compatible Transformer for Irregularly Sampled Multivariate Time Series              Yuxi Wei, Juntong Peng, Tong He, Chenxin Xu, Jian Zhang, Shirui Pan, Siheng Chen            2023-10-17 2023 IEEE International Conference on Data Mining (ICDM) 0 9 visibility_off Learning from Irregularly-Sampled Time Series: A Missing Data Perspective              Steven Cheng-Xian Li, Benjamin M Marlin            2020-07-12 ArXiv, MAG, DBLP 54 34 visibility_off Continual Learning for Multivariate Time Series Tasks with Variable Input Dimensions              Vibhor Gupta, Jyoti Narwariya, Pankaj Malhotra, L. Vig, Gautam M. Shroff            2021-12-01 2021 IEEE International Conference on Data Mining (ICDM) 17 28 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01/","title":"4971f9abd024e40fbbdff2e9492745b68a6bca01","text":""},{"location":"recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01/#_1","title":"4971f9abd024e40fbbdff2e9492745b68a6bca01","text":"<p> This page was last updated on 2024-11-11 06:05:46 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Sparse identification of Lagrangian for nonlinear dynamical systems via proximal gradient method              Adam Purnomo, M. Hayashibe            2022-09-04 Scientific Reports 3 24 visibility_off Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI)              Gabriel F. Machado, Morgan Jones            2023-10-06 2024 American Control Conference (ACC) 1 1 visibility_off PySINDy: A comprehensive Python package for robust sparse system identification              A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2021-11-12 J. Open Source Softw. 119 67 visibility_off Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian              H. Chu, M. Hayashibe            2020-01-31 IEEE Robotics and Automation Letters 24 24 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 11 2 visibility_off Derivative-Based SINDy (DSINDy): Addressing the Challenge of Discovering Governing Equations from Noisy Data              J. Wentz, A. Doostan            2022-11-10 SSRN Electronic Journal 14 33 visibility_off A Robust SINDy Approach by Combining Neural Networks and an Integral Form              Ali Forootani, P. Goyal, P. Benner            2023-09-13 ArXiv 2 15 visibility_off ADAM-SINDy: An Efficient Optimization Framework for Parameterized Nonlinear Dynamical System Identification              Siva Viknesh, Younes Tatari, Amirhossein Arzani            2024-10-21 ArXiv 0 24 visibility_off Generalizing the SINDy approach with nested neural networks              Camilla Fiorini, Cl'ement Flint, Louis Fostier, Emmanuel Franck, Reyhaneh Hashemi, Victor Michel-Dansac, Wassim Tenachi            2024-04-24 ArXiv 0 3 visibility_off SINDy vs Hard Nonlinearities and Hidden Dynamics: a Benchmarking Study              Aurelio Raffa Ugolini, Valentina Breschi, Andrea Manzoni, M. Tanelli            2024-03-01 ArXiv 1 2 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/4a111f7a3b56d0468f13104999844885157ef17d/","title":"4a111f7a3b56d0468f13104999844885157ef17d","text":""},{"location":"recommendations/4a111f7a3b56d0468f13104999844885157ef17d/#_1","title":"4a111f7a3b56d0468f13104999844885157ef17d","text":"<p> This page was last updated on 2024-11-11 06:05:17 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Unified Training of Universal Time Series Forecasting Transformers Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation              Taha \u0130brahim Aksu, Gerald Woo, Juncheng Liu, Xu Liu, Chenghao Liu, Silvio Savarese, Caiming Xiong, Doyen Sahoo            2024-10-14 ArXiv 0 25 visibility_off FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting              Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang            2024-10-15 ArXiv 0 27 visibility_off Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts              X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin            2024-09-24 ArXiv 2 9 visibility_off Chronos: Learning the Language of Time Series              Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang            2024-03-12 ArXiv 69 18 visibility_off Timer: Generative Pre-trained Transformers Are Large Time Series Models              Yong Liu, Haoran Zhang, Chenyu Li, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-02-04 ArXiv, DBLP 21 67 visibility_off A Time Series is Worth 64 Words: Long-term Forecasting with Transformers              Yuqi Nie, Nam H. Nguyen, Phanwadee Sinthong, J. Kalagnanam            2022-11-27 ArXiv 724 35 visibility_off FlexTSF: A Universal Forecasting Model for Time Series with Variable Regularities              Jingge Xiao, Yile Chen, Gao Cong, Wolfgang Nejdl, Simon Gottschalk            2024-10-30 ArXiv 0 6 visibility_off HiMTM: Hierarchical Multi-Scale Masked Time Series Modeling with Self-Distillation for Long-Term Forecasting              Shubao Zhao, Ming Jin, Zhaoxiang Hou, Che-Sheng Yang, Zengxiang Li, Qingsong Wen, Yi Wang            2024-01-10 Proceedings of the 33rd ACM International Conference on Information and Knowledge Management 0 7 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14/","title":"5bd2c0acaf58c25f71617db2396188c74d29bf14","text":""},{"location":"recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14/#_1","title":"5bd2c0acaf58c25f71617db2396188c74d29bf14","text":"<p> This page was last updated on 2024-11-11 06:05:23 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Domain Adaptation for Time Series Under Feature and Label Shifts Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Deep Unsupervised Domain Adaptation for Time Series Classification: a Benchmark              Hassan Ismail Fawaz, Ganesh Del Grosso, Tanguy Kerdoncuff, Aur\u00e9lie Boisbunon, Illyyne Saffar            2023-12-15 ArXiv 1 5 visibility_off A Virtual-Label-Based Hierarchical Domain Adaptation Method for Time-Series Classification.              Wenmian Yang, Lizhi Cheng, Mohamed Ragab, Min Wu, Sinno Jialin Pan, Zhenghua Chen            2024-08-28 IEEE transactions on neural networks and learning systems 0 15 visibility_off ADATIME: A Benchmarking Suite for Domain Adaptation on Time Series Data              Mohamed Ragab, Emadeldeen Eldele, Wee Ling Tan, Chuan-Sheng Foo, Zhenghua Chen, Min Wu, C. Kwoh, Xiaoli Li            2022-03-15 ACM Transactions on Knowledge Discovery from Data 51 39 visibility_off LogoRA: Local-Global Representation Alignment for Robust Time Series Classification              Huanyu Zhang, Yi-Fan Zhang, Zhang Zhang, Qingsong Wen, Liang Wang            2024-09-12 ArXiv 0 4 visibility_off GLA-DA: Global-Local Alignment Domain Adaptation for Multivariate Time Series              Gang Tu, Dan Li, Bingxin Lin, Zibin Zheng, See-Kiong Ng            2024-10-09 ArXiv 0 1 visibility_off Domain Generalization via Selective Consistency Regularization for Time Series Classification              Wenyu Zhang, Mohamed Ragab, Chuan-Sheng Foo            2022-06-16 2022 26th International Conference on Pattern Recognition (ICPR) 0 25 visibility_off Domain Adaptation for Time Series Forecasting via Attention Sharing              Xiaoyong Jin, Youngsuk Park, Danielle C. Maddix, Bernie Wang, Xifeng Yan            2021-02-13 ArXiv, DBLP 63 65 visibility_off Domain Adaptation with Representation Learning and Nonlinear Relation for Time Series              A. Hussein, Hazem Hajj            2022-02-15 ACM Transactions on Internet of Things 11 9 visibility_off Match-And-Deform: Time Series Domain Adaptation through Optimal Transport and Temporal Alignment              Franccois Painblanc, L. Chapel, N. Courty, Chlo\u00e9 Friguet, Charlotte Pelletier, R. Tavenard            2023-08-24 ArXiv 2 34 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/5d150cec2775f9bc863760448f14104cc8f42368/","title":"5d150cec2775f9bc863760448f14104cc8f42368","text":""},{"location":"recommendations/5d150cec2775f9bc863760448f14104cc8f42368/#_1","title":"5d150cec2775f9bc863760448f14104cc8f42368","text":"<p> This page was last updated on 2024-11-11 06:05:41 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Discovering governing equations from data by sparse identification of nonlinear dynamical systems Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Data-driven discovery of coordinates and governing equations              Kathleen P. Champion, Bethany Lusch, J. Kutz, S. Brunton            2019-03-29 Proceedings of the National Academy of Sciences of the United States of America 648 67 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 11 2 visibility_off Sparse Estimation for Hamiltonian Mechanics              Yuya Note, Masahito Watanabe, Hiroaki Yoshimura, Takaharu Yaguchi, Toshiaki Omori            2024-03-25 Mathematics 0 11 visibility_off Physics-informed learning of governing equations from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 Nature Communications 289 13 visibility_off Sparsistent Model Discovery              Georges Tod, G. Both, R. Kusters            2021-06-22 ArXiv 1 13 visibility_off Deep learning of physical laws from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 ArXiv 19 13 visibility_off Discovering Governing equations from Graph-Structured Data by Sparse Identification of Nonlinear Dynamical Systems              Mohammad Amin Basiri, Sina Khanmohammadi            2024-09-02 ArXiv 0 3 visibility_off Discovery of nonlinear dynamical systems using a Runge\u2013Kutta inspired dictionary-based sparse regression approach              P. Goyal, P. Benner            2021-05-11 Proceedings. Mathematical, Physical, and Engineering Sciences 41 53 visibility_off Exploiting sparsity and equation-free architectures in complex systems              J. Proctor, S. Brunton, Bingni W. Brunton, J. Kutz            2014-12-10 The European Physical Journal Special Topics 64 67 visibility_off Exploiting sparsity and equation-free architectures in complex systems              J. Proctor, S. Brunton, Bingni W. Brunton, J. Kutz            2014-12-01 The European Physical Journal Special Topics 7 67 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c/","title":"60d0d998fa038182b3b69a57adb9b2f82d40589c","text":""},{"location":"recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c/#_1","title":"60d0d998fa038182b3b69a57adb9b2f82d40589c","text":"<p> This page was last updated on 2024-11-11 06:05:41 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Robust learning from noisy, incomplete, high-dimensional experimental data via physically constrained symbolic regression Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Automated discovery of fundamental variables hidden in experimental data              Boyuan Chen, Kuang Huang, Sunand Raghupathi, I. Chandratreya, Qi Du, H. Lipson            2022-07-01 Nature Computational Science 77 21 visibility_off Opportunities for machine learning in scientific discovery              Ricardo Vinuesa, Jean Rabault, H. Azizpour, Stefan Bauer, Bingni W. Brunton, Arne Elofsson, Elias Jarlebring, Hedvig Kjellstrom, Stefano Markidis, David Marlevi, Paola Cinnella, S. Brunton            2024-05-07 ArXiv 1 67 visibility_off DeepMoD: Deep learning for model discovery in noisy data              G. Both, Subham Choudhury, P. Sens, R. Kusters            2019-04-20 J. Comput. Phys. 104 36 visibility_off Discovering sparse interpretable dynamics from partial observations              Peter Y. Lu, Joan Ari\u00f1o Bernad, M. Solja\u010di\u0107            2021-07-22 Communications Physics 20 95 visibility_off PNAS Plus Significance Statements              Ronald R. Coifman, David A. Kessler, A. Goodkind            2017-09-19 Proceedings of the National Academy of Sciences 40 13 visibility_off Discovering State Variables Hidden in Experimental Data              Boyuan Chen, Kuang Huang, Sunand Raghupathi, I. Chandratreya, Qi Du, Hod Lipson            2021-12-20 ArXiv 13 73 visibility_off A physics-informed operator regression framework for extracting data-driven continuum models              Ravi G. Patel, N. Trask, M. Wood, E. Cyr            2020-09-25 ArXiv 90 17 visibility_off Data-Driven Discovery of Coarse-Grained Equations              Joseph Bakarji, D. Tartakovsky            2020-01-30 J. Comput. Phys. 33 47 visibility_off Universal Differential Equations for Scientific Machine Learning              Christopher Rackauckas, Yingbo Ma, Julius Martensen, Collin Warner, K. Zubov, R. Supekar, Dominic J. Skinner, A. Ramadhan            2020-01-13 ArXiv 509 26 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37/","title":"648d90b713997a771e2c49f02cd771e8b7b10b37","text":""},{"location":"recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37/#_1","title":"648d90b713997a771e2c49f02cd771e8b7b10b37","text":"<p> This page was last updated on 2024-11-11 06:05:21 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistency Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Self-Supervised Contrastive Learning for Medical Time Series: A Systematic Review              Ziyu Liu, A. Alavi, Minyi Li, X. Zhang            2023-04-23 Sensors (Basel, Switzerland) 28 77 visibility_off Time-series representation learning via Time-Frequency Fusion Contrasting              Wenbo Zhao, Ling Fan            2024-06-12 Frontiers in Artificial Intelligence 0 1 visibility_off TS-MoCo: Time-Series Momentum Contrast for Self-Supervised Physiological Representation Learning              Philipp Hallgarten, David Bethge, Ozan \u00d6zdenizci, T. Gro\u00dfe-Puppendahl, Enkelejda Kasneci            2023-06-10 2023 31st European Signal Processing Conference (EUSIPCO) 0 39 visibility_off Supervised Contrastive Few-Shot Learning for High-Frequency Time Series              X. Chen, Cheng Ge, Ming Wang, Jin Wang            2023-06-26 DBLP 4 140 visibility_off Contrast Everything: A Hierarchical Contrastive Framework for Medical Time-Series              Yihe Wang, Yu Han, Haishuai Wang, Xiang Zhang            2023-10-21 ArXiv 30 4 visibility_off CALDA: Improving Multi-Source Time Series Domain Adaptation With Contrastive Adversarial Learning              Garrett Wilson, J. Doppa, D. Cook            2021-09-30 IEEE Transactions on Pattern Analysis and Machine Intelligence 21 72 visibility_off Self-supervised Classification of Clinical Multivariate Time Series using Time Series Dynamics              Yakir Yehuda, Daniel Freedman, Kira Radinsky            2023-08-04 Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 0 25 visibility_off Improving Time Series Encoding with Noise-Aware Self-Supervised Learning and an Efficient Encoder              Anh Duy Nguyen, Trang H. Tran, Hieu Pham, Phi-Le Nguyen, Lam M. Nguyen            2023-06-11 ArXiv 3 20 visibility_off Phase-driven Domain Generalizable Learning for Nonstationary Time Series              Payal Mohapatra, Lixu Wang, Qi Zhu            2024-02-05 ArXiv 1 10 visibility_off Contrastive Neural Processes for Self-Supervised Learning              Konstantinos Kallidromitis, Denis A. Gudovskiy, Kozuka Kazuki, Ohama Iku, Luca Rigazio            2021-10-24 ArXiv, DBLP 10 13 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/73dd9c49f205280991826b2ea4b50344203916b4/","title":"73dd9c49f205280991826b2ea4b50344203916b4","text":""},{"location":"recommendations/73dd9c49f205280991826b2ea4b50344203916b4/#_1","title":"73dd9c49f205280991826b2ea4b50344203916b4","text":"<p> This page was last updated on 2024-11-11 06:05:49 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Learning Discrepancy Models From Experimental Data Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Sparse identification of Lagrangian for nonlinear dynamical systems via proximal gradient method              Adam Purnomo, M. Hayashibe            2022-09-04 Scientific Reports 3 24 visibility_off Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian              H. Chu, M. Hayashibe            2020-01-31 IEEE Robotics and Automation Letters 24 24 visibility_off SINDy vs Hard Nonlinearities and Hidden Dynamics: a Benchmarking Study              Aurelio Raffa Ugolini, Valentina Breschi, Andrea Manzoni, M. Tanelli            2024-03-01 ArXiv 1 2 visibility_off Machine Learning and System Identification for Estimation in Physical Systems              Fredrik Bagge Carlson            2018-12-20 ArXiv 5 8 visibility_off Composed Physics- and Data-driven System Identification for Non-autonomous Systems in Control Engineering              Ricarda-Samantha G\u00f6tte, Julia Timmermann            2021-12-15 2022 3rd International Conference on Artificial Intelligence, Robotics and Control (AIRC) 5 5 visibility_off Sparse identification of nonlinear dynamics for model predictive control in the low-data limit              E. Kaiser, J. Kutz, S. Brunton            2017-11-15 Proceedings. Mathematical, Physical, and Engineering Sciences 461 67 visibility_off Data-based modeling and control of nonlinear process systems using sparse identification: An overview of recent results              Fahim Abdullah, P. Christofides            2023-03-01 Comput. Chem. Eng. 15 76 visibility_off SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics              Kadierdan Kaheman, J. Kutz, S. Brunton            2020-04-05 Proceedings. Mathematical, Physical, and Engineering Sciences 221 67 visibility_off Learning Dynamical Systems by Leveraging Data from Similar Systems              Lei Xin, Lintao Ye, G. Chiu, S. Sundaram            2023-02-08 ArXiv 7 37 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/883547fdbd88552328a6615ec620f96e39c57018/","title":"883547fdbd88552328a6615ec620f96e39c57018","text":""},{"location":"recommendations/883547fdbd88552328a6615ec620f96e39c57018/#_1","title":"883547fdbd88552328a6615ec620f96e39c57018","text":"<p> This page was last updated on 2024-11-11 06:05:51 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Ensemble-SINDy: Robust sparse model discovery in the low-data, high-noise limit, with active learning and control Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Sparsifying priors for Bayesian uncertainty quantification in model discovery              Seth M. Hirsh, D. Barajas-Solano, J. Kutz            2021-07-05 Royal Society Open Science 61 32 visibility_off Convergence of uncertainty estimates in Ensemble and Bayesian sparse model discovery              Liyao (Mars) Gao, Urban Fasel, S. Brunton, J. Kutz            2023-01-30 ArXiv 11 67 visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 11 2 visibility_off SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics              Kadierdan Kaheman, J. Kutz, S. Brunton            2020-04-05 Proceedings. Mathematical, Physical, and Engineering Sciences 221 67 visibility_off Rapid Bayesian identification of sparse nonlinear dynamics from scarce and noisy data              Lloyd Fung, Urban Fasel, M. Juniper            2024-02-23 ArXiv 1 38 visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3408 67 visibility_off Sparse identification of nonlinear dynamics in the presence of library and system uncertainty              Andrew O'Brien            2024-01-23 ArXiv 0 0 visibility_off Comparative Analysis of Uncertainty Quantification Models in Active Learning for Efficient System Identification of Dynamical Systems              Hans Mertens, Frances Zhu            2024-08-28 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE) 0 0 $\\dot { \\boldsymbol x} = { \\boldsymbol f} ({ \\boldsymbol x})$ . First, we propose, for use in high-noise settings, an extensive toolkit of critically enabling extensions for the SINDy regression method, to progressively cull functionals from an over-complete library and yield a set of sparse equations that regress to the derivate  $\\dot { \\boldsymbol {x}}$ . This toolkit includes: (regression step) weight timepoints based on estimated noise, use ensembles to estimate coefficients, and regress using FFTs; (culling step) leverage linear dependence of functionals, and restore and protect culled functionals based on Figures of Merit (FoMs). In a novel Assessment step, we define FoMs that compare model predictions to the original time-series (i.e.,  ${ \\boldsymbol x}(t)$  rather than  $\\dot { \\boldsymbol {x}}(t)$ ). These innovations can extract sparse governing equations and coefficients from high-noise time-series data (e.g., 300% added noise). For example, it discovers the correct sparse libraries in the Lorenz system, with median coefficient estimate errors equal to 1%\u22123% (for 50% noise), 6%\u22128% (for 100% noise), and 23%\u221225% (for 300% noise). The enabling modules in the toolkit are combined into a single method, but the individual modules can be tactically applied in other equation discovery methods (SINDy or not) to improve results on high-noise data. Second, we propose a technique, applicable to any model discovery method based on  $\\dot { \\boldsymbol x} = { \\boldsymbol f} ({ \\boldsymbol x})$ , to assess the accuracy of a discovered model in the context of non-unique solutions due to noisy data. Currently, this non-uniqueness can obscure a discovered model\u2019s accuracy and thus a discovery method\u2019s effectiveness. We describe a technique that uses linear dependencies among functionals to transform a discovered model into an equivalent form that is closest to the true model, enabling more accurate assessment of a discovered model\u2019s correctness.\"&gt;           visibility_off A Toolkit for Data-Driven Discovery of Governing Equations in High-Noise Regimes              Charles B. Delahunt, J. Kutz            2021-11-08 IEEE Access 16 32 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/b2eb064f432557c59ce99834d7dc7817e4687271/","title":"B2eb064f432557c59ce99834d7dc7817e4687271","text":""},{"location":"recommendations/b2eb064f432557c59ce99834d7dc7817e4687271/#_1","title":"B2eb064f432557c59ce99834d7dc7817e4687271","text":"<p> This page was last updated on 2024-11-11 06:05:43 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Sparse identification of nonlinear dynamics for model predictive control in the low-data limit Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off SINDy vs Hard Nonlinearities and Hidden Dynamics: a Benchmarking Study              Aurelio Raffa Ugolini, Valentina Breschi, Andrea Manzoni, M. Tanelli            2024-03-01 ArXiv 1 2 visibility_off Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian              H. Chu, M. Hayashibe            2020-01-31 IEEE Robotics and Automation Letters 24 24 visibility_off SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics              Kadierdan Kaheman, J. Kutz, S. Brunton            2020-04-05 Proceedings. Mathematical, Physical, and Engineering Sciences 221 67 visibility_off Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI)              Gabriel F. Machado, Morgan Jones            2023-10-06 2024 American Control Conference (ACC) 1 1 visibility_off Multi-objective SINDy for parameterized model discovery from single transient trajectory data              Javier A. Lemus, Benjamin Herrmann            2024-05-14 ArXiv 0 0 visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3408 67 visibility_off PySINDy: A comprehensive Python package for robust sparse system identification              A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2021-11-12 J. Open Source Softw. 119 67 visibility_off ADAM-SINDy: An Efficient Optimization Framework for Parameterized Nonlinear Dynamical System Identification              Siva Viknesh, Younes Tatari, Amirhossein Arzani            2024-10-21 ArXiv 0 24 visibility_off Nonlinear Control of Networked Dynamical Systems              Megan Morrison, Nathan Kutz            2020-06-09 IEEE Transactions on Network Science and Engineering 10 5 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da/","title":"C0fc3882a9976f6a9cdc3a724bce184b786503da","text":""},{"location":"recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da/#_1","title":"C0fc3882a9976f6a9cdc3a724bce184b786503da","text":"<p> This page was last updated on 2024-11-11 06:05:54 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article A Unified Framework for Sparse Relaxed Regularized Regression: SR3 Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Sparse Relaxed Regularized Regression: SR3              P. Zheng, T. Askham, S. Brunton, J. Kutz, A. Aravkin            2018-07-14 ArXiv 9 67 visibility_off Rank-one Convexification for Sparse Regression              Alper Atamt\u00fcrk, A. G\u00f3mez            2019-01-29 ArXiv 48 36 visibility_off Sparse Recovery via Partial Regularization: Models, Theory and Algorithms              Zhaosong Lu, Xiaorui Li            2015-11-23 ArXiv 37 32 visibility_off Structured Regularizers for High-Dimensional Problems: Statistical and Computational Issues              M. Wainwright            2014-01-03 59 95 visibility_off Compressed Sparse Linear Regression              S. Kasiviswanathan, M. Rudelson            2017-07-25 ArXiv 1 30 visibility_off WARPd: A linearly convergent first-order method for inverse problems with approximate sharpness conditions              Matthew J. Colbrook            2021-10-24 ArXiv 2 17 $\\ell _1$ -norm as the loss function for the residual error and utilizes a generalized nonconvex penalty for sparsity inducing. The $\\ell _1$ -loss is less sensitive to outliers in the measurements than the popular  $\\ell _2$-loss, while the nonconvex penalty has the capability of ameliorating the bias problem of the popular convex LASSO penalty and thus can yield more accurate recovery. To solve this nonconvex and nonsmooth minimization formulation efficiently, we propose a first-order algorithm based on alternating direction method of multipliers. A smoothing strategy on the $\\ell _1$ -loss function has been used in deriving the new algorithm to make it convergent. Further, a sufficient condition for the convergence of the new algorithm has been provided for generalized nonconvex regularization. In comparison with several state-of-the-art algorithms, the new algorithm showed better performance in numerical experiments in recovering sparse signals and compressible images. The new algorithm scales well for large-scale problems, as often encountered in image processing.\"&gt;           visibility_off Efficient and Robust Recovery of Sparse Signal and Image Using Generalized Nonconvex Regularization              Fei Wen, L. Pei, Yuan Yang, Wenxian Yu, Peilin Liu            2017-03-23 IEEE Transactions on Computational Imaging 87 31 visibility_off Regularizers for structured sparsity              C. Micchelli, Jean Morales, M. Pontil            2010-10-04 Advances in Computational Mathematics 80 70 visibility_off Regularizers for structured sparsity              C. Micchelli, Jean Morales, M. Pontil            2010-10-04 Advances in Computational Mathematics 80 70 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524/","title":"C3c94ccc094dcf546e8e31c9a42506302e837524","text":""},{"location":"recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524/#_1","title":"C3c94ccc094dcf546e8e31c9a42506302e837524","text":"<p> This page was last updated on 2024-11-11 06:05:25 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article AZ-whiteness test: a test for signal uncorrelation on spatio-temporal graphs Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off AZ-whiteness test: a test for uncorrelated noise on spatio-temporal graphs              Daniele Zambon, C. Alippi            2022-04-23 ArXiv 7 50 visibility_off A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection              Ming Jin, Huan Yee Koh, Qingsong Wen, Daniele Zambon, C. Alippi, G. I. Webb, Irwin King, Shirui Pan            2023-07-07 IEEE Transactions on Pattern Analysis and Machine Intelligence 82 50 visibility_off Learning Time-Aware Graph Structures for Spatially Correlated Time Series Forecasting              Minbo Ma, Jilin Hu, Christian S. Jensen, Fei Teng, Peng Han, Zhiqiang Xu, Tian-Jie Li            2023-12-27 2024 IEEE 40th International Conference on Data Engineering (ICDE) 0 13 visibility_off Graph construction on complex spatiotemporal data for enhancing graph neural network-based approaches              Stefan Bloemheuvel, J. Hoogen, Martin Atzmueller            2023-09-25 International Journal of Data Science and Analytics 2 6 visibility_off Time-Varying Signals Recovery Via Graph Neural Networks              Jhon A. Castro-Correa, Jhony H. Giraldo, Anindya Mondal, M. Badiey, T. Bouwmans, Fragkiskos D. Malliaros            2023-02-22 ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 7 42 visibility_off Graph-Time Convolutional Neural Networks: Architecture and Theoretical Analysis              Mohammad Sabbaqi, E. Isufi            2022-06-30 IEEE Transactions on Pattern Analysis and Machine Intelligence 8 22 visibility_off Multivariate Time Series Forecasting With Dynamic Graph Neural ODEs              Ming Jin, Yu Zheng, Yuanhao Li, Siheng Chen, B. Yang, Shirui Pan            2022-02-17 IEEE Transactions on Knowledge and Data Engineering 69 44 visibility_off Dynamic Graph Learning with Long and Short-Term for Multivariate Time Series Anomaly Detection              Yuyin Tian, Rong Gao, Lingyu Yan, Donghua Liu, Zhiwei Ye            2023-09-07 2023 IEEE 12th International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS) 0 4 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/ccea298edb788edf821aef58f0952c3e8debc25a/","title":"Ccea298edb788edf821aef58f0952c3e8debc25a","text":""},{"location":"recommendations/ccea298edb788edf821aef58f0952c3e8debc25a/#_1","title":"Ccea298edb788edf821aef58f0952c3e8debc25a","text":"<p> This page was last updated on 2024-11-11 06:05:11 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Graph Deep Learning for Time Series Forecasting Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting              Nancy R. Xu, Chrysoula Kosma, M. Vazirgiannis            2023-07-27 ArXiv 1 55 visibility_off Balanced Graph Structure Learning for Multivariate Time Series Forecasting              Weijun Chen, Yanze Wang, Chengshuo Du, Zhenglong Jia, Feng Liu, Ran Chen            2022-01-24 ArXiv 0 11 visibility_off Sparse Graph Learning from Spatiotemporal Time Series              Andrea Cini, Daniele Zambon, C. Alippi            2022-05-26 J. Mach. Learn. Res. 14 50 visibility_off ForecastGrapher: Redefining Multivariate Time Series Forecasting with Graph Neural Networks              Wanlin Cai, Kun Wang, Hao Wu, Xiaoxu Chen, Yuankai Wu            2024-05-28 ArXiv 0 3 visibility_off Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks              Zonghan Wu, Shirui Pan, Guodong Long, Jing Jiang, Xiaojun Chang, Chengqi Zhang            2020-05-24 Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery &amp; Data Mining 1082 57 visibility_off DeepHGNN: Study of Graph Neural Network based Forecasting Methods for Hierarchically Related Multivariate Time Series              Abishek Sriramulu, Nicolas Fourrier, Christoph Bergmeir            2024-05-29 ArXiv 0 4 visibility_off FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective              Kun Yi, Qi Zhang, Wei Fan, Hui He, Liang Hu, Pengyang Wang, Ning An, Longbin Cao, Zhendong Niu            2023-11-10 ArXiv 49 6 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9/","title":"D3dbbd0f0de51b421a6220bd6480b8d2e99a88e9","text":""},{"location":"recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9/#_1","title":"D3dbbd0f0de51b421a6220bd6480b8d2e99a88e9","text":"<p> This page was last updated on 2024-11-11 06:05:07 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Graph Time-series Modeling in Deep Learning: A Survey              Hongjie Chen, Hoda Eldardiry            2023-12-23 ACM Transactions on Knowledge Discovery from Data 3 15 visibility_off A Systematic Literature Review of Spatio-Temporal Graph Neural Network Models for Time Series Forecasting and Classification              Flavio Corradini, Marco Gori, Carlo Lucheroni, Marco Piangerelli, Martina Zannotti            2024-10-29 ArXiv 0 12 visibility_off TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting              Nancy R. Xu, Chrysoula Kosma, M. Vazirgiannis            2023-07-27 ArXiv 1 55 visibility_off Graph Deep Learning for Time Series Forecasting              Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-10-24 ArXiv 9 50 visibility_off Graph Anomaly Detection in Time Series: A Survey              Thi Kieu Khanh Ho, Ali Karami, N. Armanfard            2023-01-31 ArXiv 2 15 visibility_off Multivariate Time-Series Anomaly Detection based on Enhancing Graph Attention Networks with Topological Analysis              Zhe Liu, Xiang Huang, Jingyun Zhang, Zhifeng Hao, L. Sun, Hao Peng            2024-08-23 ArXiv 0 2 visibility_off FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective              Kun Yi, Qi Zhang, Wei Fan, Hui He, Liang Hu, Pengyang Wang, Ning An, Longbin Cao, Zhendong Niu            2023-11-10 ArXiv 49 6 visibility_off MGADN: A Multi-task Graph Anomaly Detection Network for Multivariate Time Series              Wei-Shu Xiong, Xiaochen (Michael) Sun            2022-11-22 ArXiv 1 2 visibility_off Multivariate Time Series Anomaly Detection via Dynamic Graph Forecasting              Katrina Chen, M. Feng, T. Wirjanto            2023-02-04 ArXiv 5 22 visibility_off Edge Conditional Node Update Graph Neural Network for Multi-variate Time Series Anomaly Detection              H. Jo, Seong-Whan Lee            2024-01-25 Inf. Sci. 1 1 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419/","title":"E2a83369383aff37224170c1ae3d3870d5d9e419","text":""},{"location":"recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419/#_1","title":"E2a83369383aff37224170c1ae3d3870d5d9e419","text":"<p> This page was last updated on 2024-11-11 06:05:10 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Taming Local Effects in Graph-based Spatiotemporal Forecasting Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Graph Deep Learning for Time Series Forecasting              Andrea Cini, Ivan Marisca, Daniele Zambon, C. Alippi            2023-10-24 ArXiv 9 50 visibility_off Scalable Spatiotemporal Graph Neural Networks              Andrea Cini, Ivan Marisca, F. Bianchi, C. Alippi            2022-09-14 ArXiv 34 50 visibility_off Sparse Graph Learning from Spatiotemporal Time Series              Andrea Cini, Daniele Zambon, C. Alippi            2022-05-26 J. Mach. Learn. Res. 14 50 visibility_off Unified Spatio-Temporal Graph Neural Networks: Data-Driven Modeling for Social Science              Yifan Li, Yu Lin, Y. Gao, L. Khan            2022-07-18 2022 International Joint Conference on Neural Networks (IJCNN) 0 12 visibility_off ForecastGrapher: Redefining Multivariate Time Series Forecasting with Graph Neural Networks              Wanlin Cai, Kun Wang, Hao Wu, Xiaoxu Chen, Yuankai Wu            2024-05-28 ArXiv 0 3 visibility_off FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective              Kun Yi, Qi Zhang, Wei Fan, Hui He, Liang Hu, Pengyang Wang, Ning An, Longbin Cao, Zhendong Niu            2023-11-10 ArXiv 49 6 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6/","title":"E2e1f1b8e6c1b7f4f166e15b7c674945856a51b6","text":""},{"location":"recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6/#_1","title":"E2e1f1b8e6c1b7f4f166e15b7c674945856a51b6","text":"<p> This page was last updated on 2024-11-11 06:05:19 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters              Mouxiang Chen, Lefei Shen, Zhuo Li, Xiaoyun Joy Wang, Jianling Sun, Chenghao Liu            2024-08-30 ArXiv 0 3 visibility_off Generative Pre-Trained Diffusion Paradigm for Zero-Shot Time Series Forecasting              Jiarui Yang, Tao Dai, Naiqi Li, Junxi Wu, Peiyuan Liu, Jinmin Li, Jigang Bao, Haigang Zhang, Shu-Tao Xia            2024-06-04 ArXiv 0 5 visibility_off Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction              Cheng Feng, Long Huang, Denis Krompass            2024-02-12 ArXiv 4 15 visibility_off Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting              Kashif Rasul, Arjun Ashok, Andrew Robert Williams, Arian Khorasani, George Adamopoulos, Rishika Bhagwatkar, Marin Bilovs, Hena Ghonia, N. Hassen, Anderson Schneider, Sahil Garg, Alexandre Drouin, Nicolas Chapados, Yuriy Nevmyvaka, I. Rish            2023-10-12 ArXiv 23 40 visibility_off Chronos: Learning the Language of Time Series              Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang            2024-03-12 ArXiv 69 18 visibility_off FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting              Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang            2024-10-15 ArXiv 0 27 visibility_off A Survey of Time Series Foundation Models: Generalizing Time Series Representation with Large Language Model              Jiexia Ye, Weiqi Zhang, Ke Yi, Yongzi Yu, Ziyue Li, Jia Li, F. Tsung            2024-05-03 ArXiv 7 47 visibility_off Multi-Patch Prediction: Adapting LLMs for Time Series Representation Learning              Yuxuan Bian, Xu Ju, Jiangtong Li, Zhijian Xu, Dawei Cheng, Qiang Xu            2024-02-07 ArXiv 8 8 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8/","title":"E6f0a85009481dcfd93aaa43ed3f980e5033b0d8","text":""},{"location":"recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8/#_1","title":"E6f0a85009481dcfd93aaa43ed3f980e5033b0d8","text":"<p> This page was last updated on 2024-11-11 06:05:53 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article Learning sparse nonlinear dynamics via mixed-integer optimization Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Automatically discovering ordinary differential equations from data with sparse regression              Kevin Egan, Weizhen Li, Rui Carvalho            2024-01-09 Communications Physics 11 2 visibility_off Discovering governing equations from data by sparse identification of nonlinear dynamical systems              S. Brunton, J. Proctor, J. Kutz            2015-09-11 Proceedings of the National Academy of Sciences 3408 67 visibility_off Sparse Identification of Nonlinear Dynamics with Side Information (SINDy-SI)              Gabriel F. Machado, Morgan Jones            2023-10-06 2024 American Control Conference (ACC) 1 1 visibility_off PySINDy: A comprehensive Python package for robust sparse system identification              A. Kaptanoglu, Brian M. de Silva, Urban Fasel, Kadierdan Kaheman, Jared L. Callaham, Charles B. Delahunt, Kathleen P. Champion, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2021-11-12 J. Open Source Softw. 119 67 visibility_off Sparse reconstruction of ordinary differential equations with inference              S. Venkatraman, Sumanta Basu, M. Wells            2023-08-17 ArXiv 0 40 visibility_off Physics-informed learning of governing equations from scarce data              Zhao Chen, Yang Liu, Hao Sun            2020-05-05 Nature Communications 289 13 visibility_off PySINDy: A Python package for the Sparse Identification of Nonlinear Dynamics from Data.              Brian M. de Silva, Kathleen P. Champion, M. Quade, Jean-Christophe Loiseau, J. Kutz, S. Brunton            2020-04-17 arXiv: Dynamical Systems 51 67 visibility_off Discovering Interpretable Dynamics by Sparsity Promotion on Energy and the Lagrangian              H. Chu, M. Hayashibe            2020-01-31 IEEE Robotics and Automation Letters 24 24 visibility_off Sparse learning of stochastic dynamical equations.              L. Boninsegna, F. N\u00fcske, C. Clementi            2017-12-06 The Journal of chemical physics 200 46 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f/","title":"F45f85fa1beaa795c24c4ff86f1f2deece72252f","text":""},{"location":"recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f/#_1","title":"F45f85fa1beaa795c24c4ff86f1f2deece72252f","text":"<p> This page was last updated on 2024-11-11 06:05:15 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article A decoder-only foundation model for time-series forecasting Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting              Kashif Rasul, Arjun Ashok, Andrew Robert Williams, Arian Khorasani, George Adamopoulos, Rishika Bhagwatkar, Marin Bilovs, Hena Ghonia, N. Hassen, Anderson Schneider, Sahil Garg, Alexandre Drouin, Nicolas Chapados, Yuriy Nevmyvaka, I. Rish            2023-10-12 ArXiv 23 40 visibility_off In-Context Fine-Tuning for Time-Series Foundation Models              Abhimanyu Das, Matthew Faw, Rajat Sen, Yichen Zhou            2024-10-31 ArXiv 0 14 visibility_off Are Language Models Actually Useful for Time Series Forecasting?              Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, Tom Hartvigsen            2024-06-22 ArXiv 13 3 visibility_off Fine-Tuning a Time Series Foundation Model with Wasserstein Loss              Andrei Chernov            2024-09-18 ArXiv 0 0 visibility_off Time-LLM: Time Series Forecasting by Reprogramming Large Language Models              Ming Jin, Shiyu Wang, Lintao Ma, Zhixuan Chu, James Y. Zhang, X. Shi, Pin-Yu Chen, Yuxuan Liang, Yuan-Fang Li, Shirui Pan, Qingsong Wen            2023-10-03 ArXiv 195 9 visibility_off In-context Time Series Predictor              Jiecheng Lu, Yan Sun, Shihao Yang            2024-05-23 ArXiv 2 2 visibility_off FoundTS: Comprehensive and Unified Benchmarking of Foundation Models for Time Series Forecasting              Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang            2024-10-15 ArXiv 0 27 visibility_off LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters              Ching Chang, Wenjie Peng, Tien-Fu Chen            2023-08-16 ArXiv 22 3 visibility_off Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts              X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin            2024-09-24 ArXiv 2 9 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"},{"location":"recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90/","title":"Ff1f1cf9df8c413ec7345da7604ba28597da5b90","text":""},{"location":"recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90/#_1","title":"Ff1f1cf9df8c413ec7345da7604ba28597da5b90","text":"<p> This page was last updated on 2024-11-11 06:05:30 UTC </p> <p>          Click here for a quick intro of the page! help </p> <p>      Recommendations for the article UNITS: A Unified Multi-Task Time Series Model Abstract Title Authors Publication Date Journal/ Conference Citation count Highest h-index visibility_off Timer: Generative Pre-trained Transformers Are Large Time Series Models              Yong Liu, Haoran Zhang, Chenyu Li, Xiangdong Huang, Jianmin Wang, Mingsheng Long            2024-02-04 ArXiv, DBLP 21 67 visibility_off Towards Generalisable Time Series Understanding Across Domains              Ozgun Turgut, Philip Muller, M. Menten, Daniel Rueckert            2024-10-09 ArXiv 0 15 visibility_off Large Pre-trained time series models for cross-domain Time series analysis tasks              Harshavardhan Kamarthi, B. A. Prakash            2023-11-19 ArXiv 4 8 visibility_off TOTEM: TOkenized Time Series EMbeddings for General Time Series Analysis              Sabera Talukder, Yisong Yue, Georgia Gkioxari            2024-02-26 ArXiv 7 4 visibility_off Universal Time-Series Representation Learning: A Survey              Patara Trirat, Yooju Shin, Junhyeok Kang, Youngeun Nam, Jihye Na, Minyoung Bae, Joeun Kim, Byunghyun Kim, Jae-Gil Lee            2024-01-08 ArXiv 6 7 visibility_off One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors              Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin            2023-11-24 ArXiv 4 8 visibility_off Chronos: Learning the Language of Time Series              Abdul Fatir Ansari, Lorenzo Stella, Caner Turkmen, Xiyuan Zhang, Pedro Mercado, Huibin Shen, Oleksandr Shchur, Syama Sundar Rangapuram, Sebastian Pineda Arango, Shubham Kapoor, Jasper Zschiegner, Danielle C. Maddix, Michael W. Mahoney, Kari Torkkola, Andrew Gordon Wilson, Michael Bohlke-Schneider, Yuyang Wang            2024-03-12 ArXiv 69 18 visibility_off Unified Training of Universal Time Series Forecasting Transformers              Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, Doyen Sahoo            2024-02-04 ArXiv 71 25 Abstract Title Authors Publication Date Journal/Conference Citation count Highest h-index </p>"}]}
\ No newline at end of file
diff --git a/site/sitemap.xml.gz b/site/sitemap.xml.gz
index 08ced7a4..8ed95744 100644
Binary files a/site/sitemap.xml.gz and b/site/sitemap.xml.gz differ
