diff --git a/docs/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378.md b/docs/recommendations/06a0ba437d41a7c82c08a9636a4438c1b5031378.md
index de6f7713..f6296712 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 2024-11-11 06:05:45 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:07 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2022-06-01</td>
           <td>Nonlinear Dynamics</td>
-          <td>30</td>
-          <td>92</td>
+          <td>31</td>
+          <td>93</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.">
@@ -140,13 +140,13 @@ hide:
     
         <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>
+          <td><a href="https://www.semanticscholar.org/paper/8474c9ab1318680cf674c97fe5af668db0c267cc" target='_blank'>Set-base dynamical parameter estimation and model invalidation for biochemical reaction networks</a></td>
           <td>
             Philipp Rumschinski, S. Borchers, S. Bosio, R. Weismantel, R. Findeisen
           </td>
           <td>2010-05-25</td>
           <td>BMC Systems Biology</td>
-          <td>67</td>
+          <td>68</td>
           <td>48</td>
         </tr>
     
diff --git a/docs/recommendations/0acd117521ef5aafb09fed02ab415523b330b058.md b/docs/recommendations/0acd117521ef5aafb09fed02ab415523b330b058.md
index 5c8725d7..ca25e179 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 2024-11-11 06:05:51 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:16 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -63,7 +63,7 @@ hide:
           <td>2019-09-15</td>
           <td>ArXiv</td>
           <td>9</td>
-          <td>107</td>
+          <td>108</td>
         </tr>
     
         <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.">
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3408</td>
-          <td>67</td>
+          <td>3467</td>
+          <td>68</td>
         </tr>
     
         <tr id="None">
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>289</td>
+          <td>300</td>
           <td>13</td>
         </tr>
     
@@ -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>67</td>
+          <td>68</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b.md b/docs/recommendations/0d01d21137a5af9f04e4b16a55a0f732cb8a540b.md
index 8c86caeb..45d8c888 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 2024-11-11 06:05:10 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:31 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -63,7 +63,7 @@ hide:
           <td>2023-06-29</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>7</td>
+          <td>8</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">
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>1</td>
+          <td>2</td>
           <td>55</td>
         </tr>
     
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2022-01-24</td>
           <td>ArXiv</td>
-          <td>0</td>
+          <td>1</td>
           <td>11</td>
         </tr>
     
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2022-07-01</td>
           <td>ArXiv, DBLP</td>
-          <td>44</td>
-          <td>35</td>
+          <td>45</td>
+          <td>36</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.">
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2021-01-18</td>
           <td>ArXiv</td>
-          <td>199</td>
+          <td>206</td>
           <td>38</td>
         </tr>
     
@@ -135,7 +135,7 @@ hide:
           <td>2024-07-02</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>9</td>
+          <td>10</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b.md b/docs/recommendations/123acfbccca0460171b6b06a4012dbb991cde55b.md
index e92846a4..e1f505be 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 2024-11-11 06:05:12 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:33 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>69</td>
+          <td>82</td>
           <td>18</td>
         </tr>
     
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2024-02-04</td>
           <td>ArXiv</td>
-          <td>5</td>
+          <td>6</td>
           <td>67</td>
         </tr>
     
@@ -86,10 +86,22 @@ hide:
           </td>
           <td>2024-06-22</td>
           <td>ArXiv</td>
-          <td>13</td>
+          <td>18</td>
           <td>3</td>
         </tr>
     
+        <tr id="Encoding time series into tokens and using language models for processing has been shown to substantially augment the models' ability to generalize to unseen tasks. However, existing language models for time series forecasting encounter several obstacles, including aliasing distortion and prolonged inference times, primarily due to the limitations of quantization processes and the computational demands of large models. This paper introduces Apollo-Forecast, a novel framework that tackles these challenges with two key innovations: the Anti-Aliasing Quantization Module (AAQM) and the Race Decoding (RD) technique. AAQM adeptly encodes sequences into tokens while mitigating high-frequency noise in the original signals, thus enhancing both signal fidelity and overall quantization efficiency. RD employs a draft model to enable parallel processing and results integration, which markedly accelerates the inference speed for long-term predictions, particularly in large-scale models. Extensive experiments on various real-world datasets show that Apollo-Forecast outperforms state-of-the-art methods by 35.41\% and 18.99\% in WQL and MASE metrics, respectively, in zero-shot scenarios. Furthermore, our method achieves a 1.9X-2.7X acceleration in inference speed over baseline methods.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/a09124813cb1d71da89d8e30719206c0668c5330" target='_blank'>Apollo-Forecast: Overcoming Aliasing and Inference Speed Challenges in Language Models for Time Series Forecasting</a></td>
+          <td>
+            Tianyi Yin, Jingwei Wang, Yunlong Ma, Han Wang, Chenze Wang, Yukai Zhao, Min Liu, Weiming Shen, Yufeng Chen
+          </td>
+          <td>2024-12-16</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>7</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.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/73f58b90697f957832f5090946894480849dea3a" target='_blank'>Timer: Generative Pre-trained Transformers Are Large Time Series Models</a></td>
@@ -98,7 +110,7 @@ hide:
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>21</td>
+          <td>26</td>
           <td>67</td>
         </tr>
     
@@ -110,7 +122,7 @@ hide:
           </td>
           <td>2023-10-05</td>
           <td>ArXiv</td>
-          <td>60</td>
+          <td>65</td>
           <td>5</td>
         </tr>
     
@@ -122,7 +134,7 @@ hide:
           </td>
           <td>2024-05-13</td>
           <td>2024 IEEE 40th International Conference on Data Engineering Workshops (ICDEW)</td>
-          <td>0</td>
+          <td>2</td>
           <td>7</td>
         </tr>
     
@@ -134,22 +146,10 @@ hide:
           </td>
           <td>2024-02-16</td>
           <td>ArXiv</td>
-          <td>5</td>
+          <td>7</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.">
-          <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>
-            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>2023-10-03</td>
-          <td>ArXiv</td>
-          <td>195</td>
-          <td>9</td>
-        </tr>
-    
   </tbody>
   <tfoot>
     <tr>
diff --git a/docs/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277.md b/docs/recommendations/16f01c1b3ddd0b2abd5ddfe4fdb3f74767607277.md
index 34a6a6be..bd6d20a0 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 2024-11-11 06:05:18 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:41 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2024-05-23</td>
           <td>ArXiv</td>
-          <td>0</td>
+          <td>1</td>
           <td>7</td>
         </tr>
     
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2024-02-16</td>
           <td>ArXiv</td>
-          <td>5</td>
+          <td>7</td>
           <td>8</td>
         </tr>
     
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2024-02-25</td>
           <td>ArXiv</td>
-          <td>11</td>
-          <td>8</td>
+          <td>14</td>
+          <td>9</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.">
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2022-09-20</td>
           <td>IEEE Transactions on Knowledge and Data Engineering</td>
-          <td>80</td>
-          <td>17</td>
+          <td>90</td>
+          <td>18</td>
         </tr>
     
         <tr id="Inspired by Large Language Models (LLMs), Time Series Forecasting (TSF), a long-standing task in time series analysis, is undergoing a transition towards Large Time Series Models (LTSMs), aiming to train universal transformer-based models for TSF. However, training LTSMs on heterogeneous time series data poses unique challenges, including diverse frequencies, dimensions, and patterns across datasets. Recent endeavors have studied and evaluated various design choices aimed at enhancing LTSM training and generalization capabilities, spanning pre-processing techniques, model configurations, and dataset configurations. In this work, we comprehensively analyze these design choices and aim to identify the best practices for training LTSM. Moreover, we propose \emph{time series prompt}, a novel statistical prompting strategy tailored to time series data. Furthermore, based on the observations in our analysis, we introduce \texttt{LTSM-bundle}, which bundles the best design choices we have identified. Empirical results demonstrate that \texttt{LTSM-bundle} achieves superior zero-shot and few-shot performances compared to state-of-the-art LSTMs and traditional TSF methods on benchmark datasets.">
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2024-05-03</td>
           <td>ArXiv</td>
-          <td>7</td>
+          <td>13</td>
           <td>47</td>
         </tr>
     
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2024-05-23</td>
           <td>ArXiv</td>
-          <td>2</td>
+          <td>3</td>
           <td>2</td>
         </tr>
     
@@ -134,22 +134,10 @@ hide:
           </td>
           <td>2023-08-16</td>
           <td>ArXiv</td>
-          <td>22</td>
+          <td>24</td>
           <td>3</td>
         </tr>
     
-        <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/e2b1ba7e73ca4ce0abef8c87f2c266e502b3b7b4" target='_blank'>Towards Time Series Reasoning with LLMs</a></td>
-          <td>
-            Winnie Chow, Lauren Gardiner, Haraldur T. Hallgr'imsson, Maxwell A. Xu, Shirley You Ren
-          </td>
-          <td>2024-09-17</td>
-          <td>ArXiv</td>
-          <td>2</td>
-          <td>1</td>
-        </tr>
-    
   </tbody>
   <tfoot>
     <tr>
diff --git a/docs/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f.md b/docs/recommendations/1df04f33a8ef313cc2067147dbb79c3ca7c5c99f.md
index 13d0161f..3139c08a 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 2024-11-11 06:05:13 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:35 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2024-02-13</td>
           <td>ArXiv, DBLP</td>
-          <td>39</td>
+          <td>43</td>
           <td>9</td>
         </tr>
     
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2024-06-09</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>3</td>
+          <td>4</td>
+          <td>4</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.">
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2022-10-08</td>
           <td>ArXiv</td>
-          <td>63</td>
+          <td>67</td>
           <td>19</td>
         </tr>
     
diff --git a/docs/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68.md b/docs/recommendations/279cd637b7e38bba1dd8915b5ce68cbcacecbe68.md
index d2250fd8..46d0b8b2 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 2024-11-11 06:05:28 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:49 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, MAG, DBLP</td>
-          <td>113</td>
+          <td>ArXiv, DBLP, MAG</td>
+          <td>115</td>
           <td>93</td>
         </tr>
     
@@ -70,7 +70,7 @@ hide:
           <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
+            Nanbo Li, Firas Laakom, Yucheng Xu, Wenyi Wang, Jurgen Schmidhuber
           </td>
           <td>2024-10-28</td>
           <td>ArXiv</td>
@@ -123,7 +123,7 @@ hide:
           <td>2020-01-13</td>
           <td>ArXiv</td>
           <td>8</td>
-          <td>60</td>
+          <td>61</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.">
@@ -134,7 +134,7 @@ hide:
           </td>
           <td>2024-06-03</td>
           <td>ArXiv</td>
-          <td>1</td>
+          <td>2</td>
           <td>12</td>
         </tr>
     
diff --git a/docs/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196.md b/docs/recommendations/2b2aa13d4959073f61ad70555bc8c7da7d116196.md
index cf399b69..e9fbe5b1 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 2024-11-11 06:05:47 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:11 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2021-10-18</td>
           <td>ArXiv</td>
-          <td>21</td>
+          <td>23</td>
           <td>51</td>
         </tr>
     
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2023-04-11</td>
           <td>SSRN Electronic Journal</td>
-          <td>3</td>
+          <td>4</td>
           <td>12</td>
         </tr>
     
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2020-11-22</td>
           <td>Biophysical Reviews</td>
-          <td>36</td>
+          <td>38</td>
           <td>5</td>
         </tr>
     
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2020-07-02</td>
           <td>ArXiv</td>
-          <td>136</td>
+          <td>139</td>
           <td>23</td>
         </tr>
     
@@ -135,7 +135,7 @@ hide:
           <td>2020-03-17</td>
           <td>Autom.</td>
           <td>18</td>
-          <td>4</td>
+          <td>5</td>
         </tr>
     
         <tr id="None">
@@ -158,7 +158,7 @@ hide:
           </td>
           <td>2022-06-20</td>
           <td>Philosophical transactions. Series A, Mathematical, physical, and engineering sciences</td>
-          <td>48</td>
+          <td>57</td>
           <td>33</td>
         </tr>
     
diff --git a/docs/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220.md b/docs/recommendations/35e2571c17246577e0bc1b9de57a314c3b60e220.md
index c1aa11ce..72dd0892 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 2024-11-11 06:05:50 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:14 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>103</td>
-          <td>67</td>
+          <td>106</td>
+          <td>68</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.">
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2023-05-31</td>
           <td>2023 American Control Conference (ACC)</td>
-          <td>22</td>
+          <td>24</td>
           <td>38</td>
         </tr>
     
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2021-09-30</td>
           <td>ArXiv, DBLP</td>
-          <td>5</td>
-          <td>105</td>
+          <td>6</td>
+          <td>106</td>
         </tr>
     
         <tr id="None">
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2022-02-17</td>
           <td>Communications Physics</td>
-          <td>25</td>
-          <td>31</td>
+          <td>26</td>
+          <td>33</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>93</td>
+          <td>95</td>
           <td>82</td>
         </tr>
     
@@ -152,14 +152,14 @@ hide:
     
         <tr id="None">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/2cd3a4931a71d74bcec53716855a3f465bca8e71" target='_blank'>Physics-informed regularization and structure preservation for learning stable reduced models from data with operator inference</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/55be4b71d7c8366b0f095e769f4586ce664c7415" target='_blank'>Discovering conservation laws using optimal transport and manifold learning</a></td>
           <td>
-            N. Sawant, B. Kramer, B. Peherstorfer
+            Peter Y. Lu, Rumen Dangovski, M. Soljavci'c
           </td>
-          <td>2021-07-06</td>
-          <td>ArXiv</td>
-          <td>25</td>
-          <td>28</td>
+          <td>2022-08-31</td>
+          <td>Nature Communications</td>
+          <td>12</td>
+          <td>13</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/3c9961153493370500020c81527b3548c96f81e0.md b/docs/recommendations/3c9961153493370500020c81527b3548c96f81e0.md
index 10eed999..2fe81f6e 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 2024-11-11 06:05:42 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:02 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>3408</td>
-          <td>67</td>
+          <td>3467</td>
+          <td>68</td>
         </tr>
     
         <tr id="None">
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>289</td>
+          <td>300</td>
           <td>13</td>
         </tr>
     
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2020-11-22</td>
           <td>Biophysical Reviews</td>
-          <td>36</td>
+          <td>38</td>
           <td>5</td>
         </tr>
     
@@ -99,7 +99,7 @@ hide:
           <td>2022-11-19</td>
           <td>Proceedings of the Royal Society A</td>
           <td>13</td>
-          <td>32</td>
+          <td>33</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>4</td>
+          <td>5</td>
           <td>37</td>
         </tr>
     
diff --git a/docs/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2.md b/docs/recommendations/3df50e9b73cc2937dfd651f4c3344bc99b7ed3f2.md
index c8ce8480..24b37797 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 2024-11-11 06:05:42 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:04 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2022-11-20</td>
           <td>Physical review. E</td>
-          <td>10</td>
-          <td>51</td>
+          <td>11</td>
+          <td>52</td>
         </tr>
     
         <tr id="A central challenge in data-driven model discovery is the presence of hidden, or latent, variables that are not directly measured but are dynamically important. Takens’ theorem provides conditions for when it is possible to augment partial measurements with time delayed information, resulting in an attractor that is diffeomorphic to that of the original full-state system. This diffeomorphism is typically unknown, and learning the dynamics in the embedding space has remained an open challenge for decades. Here, we design a deep autoencoder network to learn a coordinate transformation from the delay embedded space into a new space, where it is possible to represent the dynamics in a sparse, closed form. We demonstrate this approach on the Lorenz, Rössler and Lotka–Volterra systems, as well as a Lorenz analogue from a video of a chaotic waterwheel experiment. This framework combines deep learning and the sparse identification of nonlinear dynamics methods to uncover interpretable models within effective coordinates.">
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2022-01-13</td>
           <td>Proceedings of the Royal Society A</td>
-          <td>75</td>
-          <td>67</td>
+          <td>77</td>
+          <td>68</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.">
@@ -102,18 +102,6 @@ hide:
           <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.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/9fb357e57624525f27224822502061cc8fb831e1" target='_blank'>Detecting chaos in lineage-trees: A deep learning approach</a></td>
-          <td>
-            H. Rappeport, Irit Levin Reisman, Naftali Tishby, N. Balaban
-          </td>
-          <td>2021-06-08</td>
-          <td>ArXiv</td>
-          <td>3</td>
-          <td>56</td>
-        </tr>
-    
         <tr id="A generally applicable robust data-driven network modeling strategy offers rapid means to predict and control complex systems. 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 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.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/b0c3d159c14b0488299269c3fd62da3b9f608709" target='_blank'>Cluster-based network modeling—From snapshots to complex dynamical systems</a></td>
@@ -122,8 +110,8 @@ hide:
           </td>
           <td>2021-06-01</td>
           <td>Science Advances</td>
-          <td>50</td>
-          <td>50</td>
+          <td>52</td>
+          <td>51</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 +123,19 @@ hide:
           <td>2020-10-30</td>
           <td>arXiv: Data Analysis, Statistics and Probability</td>
           <td>1</td>
-          <td>50</td>
+          <td>51</td>
+        </tr>
+    
+        <tr id="Chaotic dynamical systems (DS) are ubiquitous in nature and society. Often we are interested in reconstructing such systems from observed time series for prediction or mechanistic insight, where by reconstruction we mean learning geometrical and invariant temporal properties of the system in question (like attractors). However, training reconstruction algorithms like recurrent neural networks (RNNs) on such systems by gradient-descent based techniques faces severe challenges. This is mainly due to exploding gradients caused by the exponential divergence of trajectories in chaotic systems. Moreover, for (scientific) interpretability we wish to have as low dimensional reconstructions as possible, preferably in a model which is mathematically tractable. Here we report that a surprisingly simple modification of teacher forcing leads to provably strictly all-time bounded gradients in training on chaotic systems, and, when paired with a simple architectural rearrangement of a tractable RNN design, piecewise-linear RNNs (PLRNNs), allows for faithful reconstruction in spaces of at most the dimensionality of the observed system. We show on several DS that with these amendments we can reconstruct DS better than current SOTA algorithms, in much lower dimensions. Performance differences were particularly compelling on real world data with which most other methods severely struggled. This work thus led to a simple yet powerful DS reconstruction algorithm which is highly interpretable at the same time.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/6ca393ee5afbe0e876baae5e428724ffd090efc5" target='_blank'>Generalized Teacher Forcing for Learning Chaotic Dynamics</a></td>
+          <td>
+            F. Hess, Z. Monfared, Manuela Brenner, D. Durstewitz
+          </td>
+          <td>2023-06-07</td>
+          <td>ArXiv, DBLP</td>
+          <td>23</td>
+          <td>37</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba.md b/docs/recommendations/455bfc515eb279cc09023faa1f78c6efb61224ba.md
index 3286942f..92f3622f 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 2024-11-11 06:05:08 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:28 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -42,6 +42,18 @@ hide:
   </thead>
   <tbody>
     
+        <tr id="Accurately modeling and analyzing time series data is crucial for downstream applications across various fields, including healthcare, finance, astronomy, and epidemiology. However, real-world time series often exhibit irregularities such as misaligned timestamps, missing entries, and variable sampling rates, complicating their analysis. Existing approaches often rely on imputation, which can introduce biases. A few approaches that directly model irregularity tend to focus exclusively on either capturing intra-series patterns or inter-series relationships, missing the benefits of integrating both. To this end, we present WaveGNN, a novel framework designed to directly (i.e., no imputation) embed irregularly sampled multivariate time series data for accurate predictions. WaveGNN utilizes a Transformer-based encoder to capture intra-series patterns by directly encoding the temporal dynamics of each time series. To capture inter-series relationships, WaveGNN uses a dynamic graph neural network model, where each node represents a sensor, and the edges capture the long- and short-term relationships between them. Our experimental results on real-world healthcare datasets demonstrate that WaveGNN consistently outperforms existing state-of-the-art methods, with an average relative improvement of 14.7% in F1-score when compared to the second-best baseline in cases with extreme sparsity. Our ablation studies reveal that both intra-series and inter-series modeling significantly contribute to this notable improvement.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/58504cccc55035221630bd35f09fb9db56f4f512" target='_blank'>WaveGNN: Modeling Irregular Multivariate Time Series for Accurate Predictions</a></td>
+          <td>
+            Arash Hajisafi, M. Siampou, Bita Azarijoo, Cyrus Shahabi
+          </td>
+          <td>2024-12-14</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>3</td>
+        </tr>
+    
         <tr id="Real-world applications often involve irregular time series, for which the time intervals between successive observations are non-uniform. Irregularity across multiple features in a multi-variate time series further results in a different subset of features at any given time (i.e., asynchronicity). Existing pre-training schemes for time-series, however, often assume regularity of time series and make no special treatment of irregularity. We argue that such irregularity offers insight about domain property of the data—for example, frequency of hospital visits may signal patient health condition—that can guide representation learning. In this work, we propose PrimeNet to learn a self-supervised representation for irregular multivariate time-series. Specifically, we design a time sensitive contrastive learning and data reconstruction task to pre-train a model. Irregular time-series exhibits considerable variations in sampling density over time. Hence, our triplet generation strategy follows the density of the original data points, preserving its native irregularity. Moreover, the sampling density variation over time makes data reconstruction difficult for different regions. Therefore, we design a data masking technique that always masks a constant time duration to accommodate reconstruction for regions of different sampling density. We learn with these tasks using unlabeled data to build a pre-trained model and fine-tune on a downstream task with limited labeled data, in contrast with existing fully supervised approach for irregular time-series, requiring large amounts of labeled data. Experiment results show that PrimeNet significantly outperforms state-of-the-art methods on naturally irregular and asynchronous data from Healthcare and IoT applications for several downstream tasks, including classification, interpolation, and regression.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/70e5e9bd353d1b1e54b4eebd3a67b58026a13dff" target='_blank'>PrimeNet: Pre-training for Irregular Multivariate Time Series</a></td>
@@ -50,7 +62,7 @@ hide:
           </td>
           <td>2023-06-26</td>
           <td>DBLP</td>
-          <td>12</td>
+          <td>13</td>
           <td>17</td>
         </tr>
     
@@ -86,8 +98,8 @@ hide:
           </td>
           <td>2023-03-01</td>
           <td>ArXiv</td>
-          <td>17</td>
-          <td>65</td>
+          <td>22</td>
+          <td>66</td>
         </tr>
     
         <tr id="Irregularly sampled time series are increasingly prevalent, particularly in medical domains. While various specialized methods have been developed to handle these irregularities, effectively modeling their complex dynamics and pronounced sparsity remains a challenge. This paper introduces a novel perspective by converting irregularly sampled time series into line graph images, then utilizing powerful pre-trained vision transformers for time series classification in the same way as image classification. This method not only largely simplifies specialized algorithm designs but also presents the potential to serve as a universal framework for time series modeling. Remarkably, despite its simplicity, our approach outperforms state-of-the-art specialized algorithms on several popular healthcare and human activity datasets. Especially in the rigorous leave-sensors-out setting where a portion of variables is omitted during testing, our method exhibits strong robustness against varying degrees of missing observations, achieving an impressive improvement of 42.8% in absolute F1 score points over leading specialized baselines even with half the variables masked. Code and data are available at https://github.com/Leezekun/ViTST">
@@ -98,8 +110,8 @@ hide:
           </td>
           <td>2023-03-01</td>
           <td>ArXiv</td>
-          <td>17</td>
-          <td>65</td>
+          <td>22</td>
+          <td>66</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.">
@@ -123,7 +135,7 @@ hide:
           <td>2023-10-17</td>
           <td>2023 IEEE International Conference on Data Mining (ICDM)</td>
           <td>0</td>
-          <td>9</td>
+          <td>10</td>
         </tr>
     
         <tr id="Irregularly-sampled time series occur in many domains including healthcare. They can be challenging to model because they do not naturally yield a fixed-dimensional representation as required by many standard machine learning models. In this paper, we consider irregular sampling from the perspective of missing data. We model observed irregularly-sampled time series data as a sequence of index-value pairs sampled from a continuous but unobserved function. We introduce an encoder-decoder framework for learning from such generic indexed sequences. We propose learning methods for this framework based on variational autoencoders and generative adversarial networks. For continuous irregularly-sampled time series, we introduce continuous convolutional layers that can efficiently interface with existing neural network architectures. Experiments show that our models are able to achieve competitive or better classification results on irregularly-sampled multivariate time series compared to recent RNN models while offering significantly faster training times.">
@@ -134,22 +146,10 @@ hide:
           </td>
           <td>2020-07-12</td>
           <td>ArXiv, MAG, DBLP</td>
-          <td>54</td>
+          <td>55</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.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/72846d22118ef02bff11330dd28f408fde72d98b" target='_blank'>Continual Learning for Multivariate Time Series Tasks with Variable Input Dimensions</a></td>
-          <td>
-            Vibhor Gupta, Jyoti Narwariya, Pankaj Malhotra, L. Vig, Gautam M. Shroff
-          </td>
-          <td>2021-12-01</td>
-          <td>2021 IEEE International Conference on Data Mining (ICDM)</td>
-          <td>17</td>
-          <td>28</td>
-        </tr>
-    
   </tbody>
   <tfoot>
     <tr>
diff --git a/docs/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01.md b/docs/recommendations/4971f9abd024e40fbbdff2e9492745b68a6bca01.md
index 5b5d1197..4e7df514 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 2024-11-11 06:05:46 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:09 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2022-09-04</td>
           <td>Scientific Reports</td>
-          <td>3</td>
+          <td>4</td>
           <td>24</td>
         </tr>
     
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2023-10-06</td>
           <td>2024 American Control Conference (ACC)</td>
-          <td>1</td>
+          <td>2</td>
           <td>1</td>
         </tr>
     
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>119</td>
-          <td>67</td>
+          <td>124</td>
+          <td>68</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.">
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2022-11-10</td>
           <td>SSRN Electronic Journal</td>
-          <td>14</td>
+          <td>15</td>
           <td>33</td>
         </tr>
     
@@ -135,7 +135,7 @@ hide:
           <td>2024-10-21</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>24</td>
+          <td>25</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.">
diff --git a/docs/recommendations/4a111f7a3b56d0468f13104999844885157ef17d.md b/docs/recommendations/4a111f7a3b56d0468f13104999844885157ef17d.md
index 7002d394..a7fa7d42 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 2024-11-11 06:05:17 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:39 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -42,9 +42,9 @@ 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.">
+        <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 23 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. Code, data, and the leaderboard can be found at https://github.com/SalesforceAIResearch/gift-eval .">
           <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><a href="https://www.semanticscholar.org/paper/8fd087d409dcaf31042a0096e3cb4dc157e1591d" 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>
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2024-10-15</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>27</td>
+          <td>1</td>
+          <td>28</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.">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>69</td>
+          <td>82</td>
           <td>18</td>
         </tr>
     
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>21</td>
+          <td>26</td>
           <td>67</td>
         </tr>
     
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2022-11-27</td>
           <td>ArXiv</td>
-          <td>724</td>
+          <td>773</td>
           <td>35</td>
         </tr>
     
@@ -118,12 +118,12 @@ hide:
           <td id="tag"><i class="material-icons">visibility_off</i></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>
-            Jingge Xiao, Yile Chen, Gao Cong, Wolfgang Nejdl, Simon Gottschalk
+            Jing Xiao, Yile Chen, Gao Cong, Wolfgang Nejdl, Simon Gottschalk
           </td>
           <td>2024-10-30</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>6</td>
+          <td>7</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.">
@@ -133,7 +133,7 @@ hide:
             Shubao Zhao, Ming Jin, Zhaoxiang Hou, Che-Sheng Yang, Zengxiang Li, Qingsong Wen, Yi Wang
           </td>
           <td>2024-01-10</td>
-          <td>Proceedings of the 33rd ACM International Conference on Information and Knowledge Management</td>
+          <td>ArXiv, DBLP</td>
           <td>0</td>
           <td>7</td>
         </tr>
diff --git a/docs/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14.md b/docs/recommendations/5bd2c0acaf58c25f71617db2396188c74d29bf14.md
index a8c3d1db..6ff3a448 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 2024-11-11 06:05:23 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:45 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>15</td>
+          <td>16</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.">
@@ -74,18 +74,18 @@ hide:
           </td>
           <td>2022-03-15</td>
           <td>ACM Transactions on Knowledge Discovery from Data</td>
-          <td>51</td>
-          <td>39</td>
+          <td>52</td>
+          <td>40</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.">
+        <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>IEEE Transactions on Knowledge and Data Engineering</td>
           <td>0</td>
           <td>4</td>
         </tr>
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2022-06-16</td>
           <td>2022 26th International Conference on Pattern Recognition (ICPR)</td>
-          <td>0</td>
+          <td>1</td>
           <td>25</td>
         </tr>
     
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2021-02-13</td>
           <td>ArXiv, DBLP</td>
-          <td>63</td>
           <td>65</td>
+          <td>66</td>
         </tr>
     
         <tr id="In many real-world scenarios, machine learning models fall short in prediction performance due to data characteristics changing from training on one source domain to testing on a target domain. There has been extensive research to address this problem with Domain Adaptation (DA) for learning domain invariant features. However, when considering advances for time series, those methods remain limited to the use of hard parameter sharing (HPS) between source and target models, and the use of domain adaptation objective function. To address these challenges, we propose a soft parameter sharing (SPS) DA architecture with representation learning while modeling the relation as non-linear between parameters of source and target models and modeling the adaptation loss function as the squared Maximum Mean Discrepancy (MMD). The proposed architecture advances the state-of-the-art for time series in the context of activity recognition and in fields with other modalities, where SPS has been limited to a linear relation. An additional contribution of our work is to provide a study that demonstrates the strengths and limitations of HPS versus SPS. Experiment results showed the success of the method in three domain adaptation cases of multivariate time series activity recognition with different users and sensors.">
@@ -134,7 +134,7 @@ hide:
           </td>
           <td>2022-02-15</td>
           <td>ACM Transactions on Internet of Things</td>
-          <td>11</td>
+          <td>12</td>
           <td>9</td>
         </tr>
     
@@ -146,7 +146,7 @@ hide:
           </td>
           <td>2023-08-24</td>
           <td>ArXiv</td>
-          <td>2</td>
+          <td>3</td>
           <td>34</td>
         </tr>
     
diff --git a/docs/recommendations/5d150cec2775f9bc863760448f14104cc8f42368.md b/docs/recommendations/5d150cec2775f9bc863760448f14104cc8f42368.md
index 582cde05..1fc73d74 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 2024-11-11 06:05:41 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:59 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>648</td>
-          <td>67</td>
+          <td>661</td>
+          <td>68</td>
         </tr>
     
         <tr id="None">
@@ -86,7 +86,7 @@ hide:
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>289</td>
+          <td>300</td>
           <td>13</td>
         </tr>
     
@@ -147,7 +147,7 @@ hide:
           <td>2014-12-10</td>
           <td>The European Physical Journal Special Topics</td>
           <td>64</td>
-          <td>67</td>
+          <td>68</td>
         </tr>
     
         <tr id="None">
@@ -159,7 +159,7 @@ hide:
           <td>2014-12-01</td>
           <td>The European Physical Journal Special Topics</td>
           <td>7</td>
-          <td>67</td>
+          <td>68</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c.md b/docs/recommendations/60d0d998fa038182b3b69a57adb9b2f82d40589c.md
index 2c0aec77..18c57be1 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 2024-11-11 06:05:41 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:01 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2022-07-01</td>
           <td>Nature Computational Science</td>
-          <td>77</td>
+          <td>78</td>
           <td>21</td>
         </tr>
     
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2024-05-07</td>
           <td>ArXiv</td>
-          <td>1</td>
-          <td>67</td>
+          <td>0</td>
+          <td>68</td>
         </tr>
     
         <tr id="None">
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2019-04-20</td>
           <td>J. Comput. Phys.</td>
-          <td>104</td>
+          <td>106</td>
           <td>36</td>
         </tr>
     
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2021-12-20</td>
           <td>ArXiv</td>
-          <td>13</td>
+          <td>14</td>
           <td>73</td>
         </tr>
     
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2020-09-25</td>
           <td>ArXiv</td>
-          <td>90</td>
-          <td>17</td>
+          <td>93</td>
+          <td>18</td>
         </tr>
     
         <tr id="None">
@@ -147,7 +147,7 @@ hide:
           </td>
           <td>2020-01-13</td>
           <td>ArXiv</td>
-          <td>509</td>
+          <td>516</td>
           <td>26</td>
         </tr>
     
diff --git a/docs/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37.md b/docs/recommendations/648d90b713997a771e2c49f02cd771e8b7b10b37.md
index 87487594..ab47eec7 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 2024-11-11 06:05:21 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:44 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,8 +50,8 @@ hide:
           </td>
           <td>2023-04-23</td>
           <td>Sensors (Basel, Switzerland)</td>
-          <td>28</td>
-          <td>77</td>
+          <td>32</td>
+          <td>78</td>
         </tr>
     
         <tr id="Time series is a typical data type in numerous domains; however, labeling large amounts of time series data can be costly and time-consuming. Learning effective representation from unlabeled time series data is a challenging task. Contrastive learning stands out as a promising method to acquire representations of unlabeled time series data. Therefore, we propose a self-supervised time-series representation learning framework via Time-Frequency Fusion Contrasting (TF-FC) to learn time-series representation from unlabeled data. Specifically, TF-FC combines time-domain augmentation with frequency-domain augmentation to generate the diverse samples. For time-domain augmentation, the raw time series data pass through the time-domain augmentation bank (such as jitter, scaling, permutation, and masking) and get time-domain augmentation data. For frequency-domain augmentation, first, the raw time series undergoes conversion into frequency domain data following Fast Fourier Transform (FFT) analysis. Then, the frequency data passes through the frequency-domain augmentation bank (such as low pass filter, remove frequency, add frequency, and phase shift) and gets frequency-domain augmentation data. The fusion method of time-domain augmentation data and frequency-domain augmentation data is kernel PCA, which is useful for extracting nonlinear features in high-dimensional spaces. By capturing both the time and frequency domains of the time series, the proposed approach is able to extract more informative features from the data, enhancing the model's capacity to distinguish between different time series. To verify the effectiveness of the TF-FC method, we conducted experiments on four time series domain datasets (i.e., SleepEEG, HAR, Gesture, and Epilepsy). Experimental results show that TF-FC significantly improves in recognition accuracy compared with other SOTA methods.">
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2024-06-12</td>
           <td>Frontiers in Artificial Intelligence</td>
-          <td>0</td>
+          <td>1</td>
           <td>1</td>
         </tr>
     
@@ -78,6 +78,18 @@ hide:
           <td>39</td>
         </tr>
     
+        <tr id="In this study, we propose a novel framework for time-series representation learning that integrates a learnable masking-augmentation strategy into a contrastive learning framework. Time-series data pose challenges due to their temporal dependencies and feature-extraction complexities. To address these challenges, we introduce a masking-based reconstruction approach within a contrastive learning context, aiming to enhance the model’s ability to learn discriminative temporal features. Our method leverages self-supervised learning to effectively capture both global and local patterns by strategically masking segments of the time-series data and reconstructing them, which aids in revealing nuanced temporal dependencies. We utilize learnable masking as a dynamic augmentation technique, which enables the model to optimize contextual relationships in the data and extract meaningful representations that are both context-aware and robust. Extensive experiments were conducted on multiple time-series datasets, including SleepEDF-78, 20, UCI-HAR, achieving improvements of 2%, 2.55%, and 3.89% each and similar performance on Epilepsy in accuracy over baseline methods. Our results show significant performance gains compared to existing methods, highlighting the potential of our framework to advance the field of time-series analysis by improving the quality of learned representations and enhancing downstream task performance.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/77dcbf1b1d32d371861cbc7b2429f37514c961c3" target='_blank'>Time-Series Representation Feature Refinement with a Learnable Masking Augmentation Framework in Contrastive Learning</a></td>
+          <td>
+            Junyeop Lee, Insung Ham, Yongmin Kim, Hanseok Ko
+          </td>
+          <td>2024-12-11</td>
+          <td>Sensors</td>
+          <td>0</td>
+          <td>2</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.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/a2056f5affe509b50e41612057ca9cca143ef97a" target='_blank'>Supervised Contrastive Few-Shot Learning for High-Frequency Time Series</a></td>
@@ -87,7 +99,7 @@ hide:
           <td>2023-06-26</td>
           <td>DBLP</td>
           <td>4</td>
-          <td>140</td>
+          <td>141</td>
         </tr>
     
         <tr id="Contrastive representation learning is crucial in medical time series analysis as it alleviates dependency on labor-intensive, domain-specific, and scarce expert annotations. However, existing contrastive learning methods primarily focus on one single data level, which fails to fully exploit the intricate nature of medical time series. To address this issue, we present COMET, an innovative hierarchical framework that leverages data consistencies at all inherent levels in medical time series. Our meticulously designed model systematically captures data consistency from four potential levels: observation, sample, trial, and patient levels. By developing contrastive loss at multiple levels, we can learn effective representations that preserve comprehensive data consistency, maximizing information utilization in a self-supervised manner. We conduct experiments in the challenging patient-independent setting. We compare COMET against six baselines using three diverse datasets, which include ECG signals for myocardial infarction and EEG signals for Alzheimer's and Parkinson's diseases. The results demonstrate that COMET consistently outperforms all baselines, particularly in setup with 10% and 1% labeled data fractions across all datasets. These results underscore the significant impact of our framework in advancing contrastive representation learning techniques for medical time series. The source code is available at https://github.com/DL4mHealth/COMET.">
@@ -98,7 +110,7 @@ hide:
           </td>
           <td>2023-10-21</td>
           <td>ArXiv</td>
-          <td>30</td>
+          <td>33</td>
           <td>4</td>
         </tr>
     
@@ -111,7 +123,7 @@ hide:
           <td>2021-09-30</td>
           <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
           <td>21</td>
-          <td>72</td>
+          <td>71</td>
         </tr>
     
         <tr id="To improve the accuracy of clinical multivariate time series (MTS) classification (such as EEG and ECG) by a novel self-supervised paradigm that directly captures the dynamics between the different time series learned together to optimize the classification task. Labels in clinical datasets are very often insufficient. One way to address this challenge is leveraging self-supervision. This paradigm attempts to identify a supervisory signal inherent within a dataset to serve as a surrogate label. We present a novel form of self-supervision: dynamics of clinical MTS. Unlike other self-supervision methods, such as masking, that are intuitive but still heuristic, we suggest to learn a representation justified by Koopman theory. The latter was shown useful for representing clinical time series and can be used as a form of surrogate task to improve the clinical MTS classification. In the ECG task, we show that our proposed framework achieved higher sensitivity and specificity than the state-of-the-art (SOTA) baseline over numerous common diagnoses. For EEG abnormality classification, our proposed framework also achieved higher sensitivity and specificity than the SOTA baseline. All results are statistically significant. Our technique yields reliable clinical diagnosis in an empirical study employing signals from thousands of patients in multiple clinical tasks employing two types of clinical-grade sensors (ECG and EEG) as compared to the state-of-the-art machine learning. Leveraging time-series-dynamics self-supervision can help mitigate the lack of labels in clinical datasets used for training machine learning algorithms and significantly improve their performance. Specifically, the ECG system presented in this work is being trialed in hospitals, used by top cardiologists for patient diagnosis and treatment. We believe that the deployment of such cutting-edge technology will significantly improve the accuracy and speed of cardiac assessments.">
@@ -150,18 +162,6 @@ hide:
           <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.">
-          <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/3f20f3b72be3b57f375bc5116a34f750554ec2a2" target='_blank'>Contrastive Neural Processes for Self-Supervised Learning</a></td>
-          <td>
-            Konstantinos Kallidromitis, Denis A. Gudovskiy, Kozuka Kazuki, Ohama Iku, Luca Rigazio
-          </td>
-          <td>2021-10-24</td>
-          <td>ArXiv, DBLP</td>
-          <td>10</td>
-          <td>13</td>
-        </tr>
-    
   </tbody>
   <tfoot>
     <tr>
diff --git a/docs/recommendations/73dd9c49f205280991826b2ea4b50344203916b4.md b/docs/recommendations/73dd9c49f205280991826b2ea4b50344203916b4.md
index 8dcb7561..0a4418b8 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 2024-11-11 06:05:49 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:12 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2022-09-04</td>
           <td>Scientific Reports</td>
-          <td>3</td>
+          <td>4</td>
           <td>24</td>
         </tr>
     
@@ -110,8 +110,8 @@ hide:
           </td>
           <td>2017-11-15</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>461</td>
-          <td>67</td>
+          <td>468</td>
+          <td>68</td>
         </tr>
     
         <tr id="None">
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2023-03-01</td>
           <td>Comput. Chem. Eng.</td>
-          <td>15</td>
+          <td>17</td>
           <td>76</td>
         </tr>
     
@@ -134,20 +134,20 @@ hide:
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>221</td>
-          <td>67</td>
+          <td>225</td>
+          <td>68</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.">
+        <tr id="Physics-based and first-principles models pervade the engineering and physical sciences, allowing for the ability to model the dynamics of complex systems with a prescribed accuracy. The approximations used in deriving governing equations often result in discrepancies between the model and sensor-based measurements of the system, revealing the approximate nature of the equations and/or the signal-to-noise ratio of the sensor itself. In modern dynamical systems, such discrepancies between model and measurement can lead to poor quantification, often undermining the ability to produce accurate and precise control algorithms. We introduce a discrepancy modeling framework to identify the missing physics and resolve the model-measurement mismatch with two distinct approaches: (i) by learning a model for the evolution of systematic state-space residual, and (ii) by discovering a model for the deterministic dynamical error. Regardless of approach, a common suite of data-driven model discovery methods can be used. The choice of method depends on one's intent (e.g., mechanistic interpretability) for discrepancy modeling, sensor measurement characteristics (e.g., quantity, quality, resolution), and constraints imposed by practical applications (e.g., modeling approaches using the suite of data-driven modeling methods on three continuous dynamical systems under varying signal-to-noise ratios. Finally, we emphasize structural shortcomings of each discrepancy modeling approach depending on error type. In summary, if the true dynamics are unknown (i.e., an imperfect model), one should learn a discrepancy model of the missing physics in the dynamical space. Yet, if the true dynamics are known yet model-measurement mismatch still exists, one should learn a discrepancy model in the state space.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/f22b5a826ef61edb741f547b6e73999476cd695d" target='_blank'>Learning Dynamical Systems by Leveraging Data from Similar Systems</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/c4a38c94d23389326f5043d63b9f5f6f0d8c0c50" target='_blank'>Discrepancy Modeling Framework: Learning Missing Physics, Modeling Systematic Residuals, and Disambiguating between Deterministic and Random Effects</a></td>
           <td>
-            Lei Xin, Lintao Ye, G. Chiu, S. Sundaram
+            Megan R. Ebers, K. Steele, J. Kutz
           </td>
-          <td>2023-02-08</td>
-          <td>ArXiv</td>
-          <td>7</td>
-          <td>37</td>
+          <td>2022-03-10</td>
+          <td>SIAM J. Appl. Dyn. Syst.</td>
+          <td>11</td>
+          <td>33</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/883547fdbd88552328a6615ec620f96e39c57018.md b/docs/recommendations/883547fdbd88552328a6615ec620f96e39c57018.md
index 4c26a2c8..57a15c76 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 2024-11-11 06:05:51 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:18 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>61</td>
-          <td>32</td>
+          <td>62</td>
+          <td>33</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>67</td>
+          <td>68</td>
         </tr>
     
         <tr id="None">
@@ -78,6 +78,18 @@ hide:
           <td>2</td>
         </tr>
     
+        <tr id="Sparse Identification of Nonlinear Dynamics (SINDy) is a data-driven algorithm for interpretability mod-eling. SINDy has a strong ability to learn dynamical systems from time series data. Sparse Identification of Nonlinear Dynamics (SINDy) can extract equations from data, but its performance will degrade when data has noise or sparse. In order to deal with this case, the idea of Monte Carlo Markov Chain (MCMC) is introduced, which is used to optimally select models. The SINDy with MCMC (MCMC-SINDy) can comprehensively discover model structures by employing random samples in data space. Simulation results show that MCMC-SINDy is more robust than SINDy with 1 % -15 % levels of noise or sparse 0.5. These results are promising for the discovery of governing equations for complex systems in neuroscience, power grids, epidemiology, finance or ecology, where governing equations have remained elusive. mportantly, compared to existing machine learning methods, the accuracy of modeling using MCMC SINDy is higher.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/6532648c8fe05d1c347e504988f1cfe4613bf71e" target='_blank'>Data-Driven Discovery of Nonlinear Dynamical Systems from Noisy and Sparse Observations</a></td>
+          <td>
+            Wei Zhu, Chao Pei, Yulan Liang, Zhang Chen, Jingsui Li
+          </td>
+          <td>2024-10-18</td>
+          <td>2024 International Conference on New Trends in Computational Intelligence (NTCI)</td>
+          <td>0</td>
+          <td>1</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.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/4971f9abd024e40fbbdff2e9492745b68a6bca01" target='_blank'>SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics</a></td>
@@ -86,8 +98,8 @@ hide:
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>221</td>
-          <td>67</td>
+          <td>225</td>
+          <td>68</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 +122,8 @@ hide:
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3408</td>
-          <td>67</td>
+          <td>3467</td>
+          <td>68</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.">
@@ -126,18 +138,6 @@ 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>
@@ -146,8 +146,8 @@ hide:
           </td>
           <td>2021-11-08</td>
           <td>IEEE Access</td>
-          <td>16</td>
-          <td>32</td>
+          <td>17</td>
+          <td>33</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271.md b/docs/recommendations/b2eb064f432557c59ce99834d7dc7817e4687271.md
index 65086187..c1baeaa9 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 2024-11-11 06:05:43 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:06 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -42,6 +42,18 @@ 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>56</td>
+          <td>68</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>
@@ -66,6 +78,18 @@ hide:
           <td>24</td>
         </tr>
     
+        <tr id="System identification, the process of deriving mathematical models of dynamical systems from observed input-output data, has undergone a paradigm shift with the advent of learning-based methods. Addressing the intricate challenges of data-driven discovery in nonlinear dynamical systems, these methods have garnered significant attention. Among them, Sparse Identification of Nonlinear Dynamics (SINDy) has emerged as a transformative approach, distilling complex dynamical behaviors into interpretable linear combinations of basis functions. However, SINDy relies on domain-specific expertise to construct its foundational"library"of basis functions, which limits its adaptability and universality. In this work, we introduce a nonlinear system identification framework called LeARN that transcends the need for prior domain knowledge by learning the library of basis functions directly from data. To enhance adaptability to evolving system dynamics under varying noise conditions, we employ a novel meta-learning-based system identification approach that uses a lightweight deep neural network (DNN) to dynamically refine these basis functions. This not only captures intricate system behaviors but also adapts seamlessly to new dynamical regimes. We validate our framework on the Neural Fly dataset, showcasing its robust adaptation and generalization capabilities. Despite its simplicity, our LeARN achieves competitive dynamical error performance compared to SINDy. This work presents a step toward the autonomous discovery of dynamical systems, paving the way for a future where machine learning uncovers the governing principles of complex systems without requiring extensive domain-specific interventions.">
+          <td id="tag"><i class="material-icons">visibility_off</i></td>
+          <td><a href="https://www.semanticscholar.org/paper/a5f516e38a88ff64971e8e5ea0c1a91ef03c8a34" target='_blank'>LeARN: Learnable and Adaptive Representations for Nonlinear Dynamics in System Identification</a></td>
+          <td>
+            Arunabh Singh, Joyjit Mukherjee
+          </td>
+          <td>2024-12-16</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>0</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.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/4971f9abd024e40fbbdff2e9492745b68a6bca01" target='_blank'>SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics</a></td>
@@ -74,8 +98,8 @@ hide:
           </td>
           <td>2020-04-05</td>
           <td>Proceedings. Mathematical, Physical, and Engineering Sciences</td>
-          <td>221</td>
-          <td>67</td>
+          <td>225</td>
+          <td>68</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.">
@@ -86,7 +110,7 @@ hide:
           </td>
           <td>2023-10-06</td>
           <td>2024 American Control Conference (ACC)</td>
-          <td>1</td>
+          <td>2</td>
           <td>1</td>
         </tr>
     
@@ -110,8 +134,8 @@ hide:
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3408</td>
-          <td>67</td>
+          <td>3467</td>
+          <td>68</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.">
@@ -122,32 +146,8 @@ hide:
           </td>
           <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>119</td>
-          <td>67</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="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/4b849d45b437a09d1ea5321a304c5bc411bf3de4" target='_blank'>Nonlinear Control of Networked Dynamical Systems</a></td>
-          <td>
-            Megan Morrison, Nathan Kutz
-          </td>
-          <td>2020-06-09</td>
-          <td>IEEE Transactions on Network Science and Engineering</td>
-          <td>10</td>
-          <td>5</td>
+          <td>124</td>
+          <td>68</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da.md b/docs/recommendations/c0fc3882a9976f6a9cdc3a724bce184b786503da.md
index 052c808a..98972c84 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 2024-11-11 06:05:54 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:21 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>67</td>
+          <td>68</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.">
@@ -87,7 +87,7 @@ hide:
           <td>2014-01-03</td>
           <td></td>
           <td>59</td>
-          <td>95</td>
+          <td>96</td>
         </tr>
     
         <tr id="High-dimensional sparse linear regression is a basic problem in machine learning and statistics. Consider a linear model $y = X\theta^\star + w$, where $y \in \mathbb{R}^n$ is the vector of observations, $X \in \mathbb{R}^{n \times d}$ is the covariate matrix and $w \in \mathbb{R}^n$ is an unknown noise vector. In many applications, the linear regression model is high-dimensional in nature, meaning that the number of observations $n$ may be substantially smaller than the number of covariates $d$. In these cases, it is common to assume that $\theta^\star$ is sparse, and the goal in sparse linear regression is to estimate this sparse $\theta^\star$, given $(X,y)$. 
@@ -125,7 +125,7 @@ As a simple consequence, our approach also provides a new way for estimating $\t
           <td>2017-03-23</td>
           <td>IEEE Transactions on Computational Imaging</td>
           <td>87</td>
-          <td>31</td>
+          <td>32</td>
         </tr>
     
         <tr id="None">
diff --git a/docs/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524.md b/docs/recommendations/c3c94ccc094dcf546e8e31c9a42506302e837524.md
index c3afc5e2..0798ebcc 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 2024-11-11 06:05:25 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:47 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2023-07-07</td>
           <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
-          <td>82</td>
+          <td>91</td>
           <td>50</td>
         </tr>
     
@@ -74,8 +74,8 @@ hide:
           </td>
           <td>2023-12-27</td>
           <td>2024 IEEE 40th International Conference on Data Engineering (ICDE)</td>
-          <td>0</td>
-          <td>13</td>
+          <td>1</td>
+          <td>14</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>2</td>
+          <td>3</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>7</td>
+          <td>9</td>
           <td>42</td>
         </tr>
     
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2022-06-30</td>
           <td>IEEE Transactions on Pattern Analysis and Machine Intelligence</td>
-          <td>8</td>
+          <td>10</td>
           <td>22</td>
         </tr>
     
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2022-02-17</td>
           <td>IEEE Transactions on Knowledge and Data Engineering</td>
-          <td>69</td>
+          <td>77</td>
           <td>44</td>
         </tr>
     
diff --git a/docs/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a.md b/docs/recommendations/ccea298edb788edf821aef58f0952c3e8debc25a.md
index 54452b13..d2e05945 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 2024-11-11 06:05:11 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:32 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>1</td>
+          <td>2</td>
           <td>55</td>
         </tr>
     
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2022-01-24</td>
           <td>ArXiv</td>
-          <td>0</td>
+          <td>1</td>
           <td>11</td>
         </tr>
     
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2020-05-24</td>
           <td>Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining</td>
-          <td>1082</td>
+          <td>1140</td>
           <td>57</td>
         </tr>
     
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2023-11-10</td>
           <td>ArXiv</td>
-          <td>49</td>
+          <td>62</td>
           <td>6</td>
         </tr>
     
diff --git a/docs/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9.md b/docs/recommendations/d3dbbd0f0de51b421a6220bd6480b8d2e99a88e9.md
index de19de59..f1c08b44 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 2024-11-11 06:05:07 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:26 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2023-12-23</td>
           <td>ACM Transactions on Knowledge Discovery from Data</td>
-          <td>3</td>
+          <td>5</td>
           <td>15</td>
         </tr>
     
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2024-10-29</td>
           <td>ArXiv</td>
-          <td>0</td>
+          <td>1</td>
           <td>12</td>
         </tr>
     
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2023-07-27</td>
           <td>ArXiv</td>
-          <td>1</td>
+          <td>2</td>
           <td>55</td>
         </tr>
     
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2023-01-31</td>
           <td>ArXiv</td>
-          <td>2</td>
+          <td>3</td>
           <td>15</td>
         </tr>
     
@@ -109,7 +109,7 @@ hide:
             Zhe Liu, Xiang Huang, Jingyun Zhang, Zhifeng Hao, L. Sun, Hao Peng
           </td>
           <td>2024-08-23</td>
-          <td>ArXiv</td>
+          <td>ArXiv, DBLP</td>
           <td>0</td>
           <td>2</td>
         </tr>
@@ -122,7 +122,7 @@ hide:
           </td>
           <td>2023-11-10</td>
           <td>ArXiv</td>
-          <td>49</td>
+          <td>62</td>
           <td>6</td>
         </tr>
     
diff --git a/docs/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419.md b/docs/recommendations/e2a83369383aff37224170c1ae3d3870d5d9e419.md
index 3833a218..dc1c46c8 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 2024-11-11 06:05:10 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:29 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -62,7 +62,7 @@ hide:
           </td>
           <td>2022-09-14</td>
           <td>ArXiv</td>
-          <td>34</td>
+          <td>37</td>
           <td>50</td>
         </tr>
     
@@ -90,28 +90,28 @@ hide:
           <td>12</td>
         </tr>
     
-        <tr id="The challenge of effectively learning inter-series correlations for multivariate time series forecasting remains a substantial and unresolved problem. Traditional deep learning models, which are largely dependent on the Transformer paradigm for modeling long sequences, often fail to integrate information from multiple time series into a coherent and universally applicable model. To bridge this gap, our paper presents ForecastGrapher, a framework reconceptualizes multivariate time series forecasting as a node regression task, providing a unique avenue for capturing the intricate temporal dynamics and inter-series correlations. Our approach is underpinned by three pivotal steps: firstly, generating custom node embeddings to reflect the temporal variations within each series; secondly, constructing an adaptive adjacency matrix to encode the inter-series correlations; and thirdly, augmenting the GNNs' expressive power by diversifying the node feature distribution. To enhance this expressive power, we introduce the Group Feature Convolution GNN (GFC-GNN). This model employs a learnable scaler to segment node features into multiple groups and applies one-dimensional convolutions with different kernel lengths to each group prior to the aggregation phase. Consequently, the GFC-GNN method enriches the diversity of node feature distribution in a fully end-to-end fashion. Through extensive experiments and ablation studies, we show that ForecastGrapher surpasses strong baselines and leading published techniques in the domain of multivariate time series forecasting.">
+        <tr id="Spatial-temporal graphs are widely used in a variety of real-world applications. Spatial-Temporal Graph Neural Networks (STGNNs) have emerged as a powerful tool to extract meaningful insights from this data. However, in real-world applications, most nodes may not possess any available temporal data during training. For example, the pandemic dynamics of most cities on a geographical graph may not be available due to the asynchronous nature of outbreaks. Such a phenomenon disagrees with the training requirements of most existing spatial-temporal forecasting methods, which jeopardizes their effectiveness and thus blocks broader deployment. In this paper, we propose to formulate a novel problem of inductive forecasting with limited training data. In particular, given a spatial-temporal graph, we aim to learn a spatial-temporal forecasting model that can be easily generalized onto those nodes without any available temporal training data. To handle this problem, we propose a principled framework named ST-FiT. ST-FiT consists of two key learning components: temporal data augmentation and spatial graph topology learning. With such a design, ST-FiT can be used on top of any existing STGNNs to achieve superior performance on the nodes without training data. Extensive experiments verify the effectiveness of ST-FiT in multiple key perspectives.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/60560da7e2483d4f788c228ebd5e226ae1f40002" target='_blank'>ForecastGrapher: Redefining Multivariate Time Series Forecasting with Graph Neural Networks</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/d05151eeedb6e1f5b10ce6c5dbd3de96467f481b" target='_blank'>ST-FiT: Inductive Spatial-Temporal Forecasting with Limited Training Data</a></td>
           <td>
-            Wanlin Cai, Kun Wang, Hao Wu, Xiaoxu Chen, Yuankai Wu
+            Zhenyu Lei, Yushun Dong, Jundong Li, Chen Chen
           </td>
-          <td>2024-05-28</td>
+          <td>2024-12-14</td>
           <td>ArXiv</td>
           <td>0</td>
-          <td>3</td>
+          <td>15</td>
         </tr>
     
-        <tr id="Multivariate time series (MTS) forecasting has shown great importance in numerous industries. Current state-of-the-art graph neural network (GNN)-based forecasting methods usually require both graph networks (e.g., GCN) and temporal networks (e.g., LSTM) to capture inter-series (spatial) dynamics and intra-series (temporal) dependencies, respectively. However, the uncertain compatibility of the two networks puts an extra burden on handcrafted model designs. Moreover, the separate spatial and temporal modeling naturally violates the unified spatiotemporal inter-dependencies in real world, which largely hinders the forecasting performance. To overcome these problems, we explore an interesting direction of directly applying graph networks and rethink MTS forecasting from a pure graph perspective. We first define a novel data structure, hypervariate graph, which regards each series value (regardless of variates or timestamps) as a graph node, and represents sliding windows as space-time fully-connected graphs. This perspective considers spatiotemporal dynamics unitedly and reformulates classic MTS forecasting into the predictions on hypervariate graphs. Then, we propose a novel architecture Fourier Graph Neural Network (FourierGNN) by stacking our proposed Fourier Graph Operator (FGO) to perform matrix multiplications in Fourier space. FourierGNN accommodates adequate expressiveness and achieves much lower complexity, which can effectively and efficiently accomplish the forecasting. Besides, our theoretical analysis reveals FGO's equivalence to graph convolutions in the time domain, which further verifies the validity of FourierGNN. Extensive experiments on seven datasets have demonstrated our superior performance with higher efficiency and fewer parameters compared with state-of-the-art methods.">
+        <tr id="The challenge of effectively learning inter-series correlations for multivariate time series forecasting remains a substantial and unresolved problem. Traditional deep learning models, which are largely dependent on the Transformer paradigm for modeling long sequences, often fail to integrate information from multiple time series into a coherent and universally applicable model. To bridge this gap, our paper presents ForecastGrapher, a framework reconceptualizes multivariate time series forecasting as a node regression task, providing a unique avenue for capturing the intricate temporal dynamics and inter-series correlations. Our approach is underpinned by three pivotal steps: firstly, generating custom node embeddings to reflect the temporal variations within each series; secondly, constructing an adaptive adjacency matrix to encode the inter-series correlations; and thirdly, augmenting the GNNs' expressive power by diversifying the node feature distribution. To enhance this expressive power, we introduce the Group Feature Convolution GNN (GFC-GNN). This model employs a learnable scaler to segment node features into multiple groups and applies one-dimensional convolutions with different kernel lengths to each group prior to the aggregation phase. Consequently, the GFC-GNN method enriches the diversity of node feature distribution in a fully end-to-end fashion. Through extensive experiments and ablation studies, we show that ForecastGrapher surpasses strong baselines and leading published techniques in the domain of multivariate time series forecasting.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/9dbcb0893d05e6176353eee401afa4929b570cf6" target='_blank'>FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/60560da7e2483d4f788c228ebd5e226ae1f40002" target='_blank'>ForecastGrapher: Redefining Multivariate Time Series Forecasting with Graph Neural Networks</a></td>
           <td>
-            Kun Yi, Qi Zhang, Wei Fan, Hui He, Liang Hu, Pengyang Wang, Ning An, Longbin Cao, Zhendong Niu
+            Wanlin Cai, Kun Wang, Hao Wu, Xiaoxu Chen, Yuankai Wu
           </td>
-          <td>2023-11-10</td>
+          <td>2024-05-28</td>
           <td>ArXiv</td>
-          <td>49</td>
-          <td>6</td>
+          <td>0</td>
+          <td>3</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6.md b/docs/recommendations/e2e1f1b8e6c1b7f4f166e15b7c674945856a51b6.md
index 9dcf24a0..39e55363 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 2024-11-11 06:05:19 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:42 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -54,6 +54,18 @@ hide:
           <td>3</td>
         </tr>
     
+        <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/30abc3fad3c88886c189e216c8414b18c396c6bd" target='_blank'>A Mamba Foundation Model for Time Series Forecasting</a></td>
+          <td>
+            Haoyu Ma, Yushu Chen, Wenlai Zhao, Jinzhe Yang, Yingsheng Ji, Xinghua Xu, Xiaozhu Liu, Hao Jing, Shengzhuo Liu, Guangwen Yang
+          </td>
+          <td>2024-11-05</td>
+          <td>ArXiv</td>
+          <td>0</td>
+          <td>8</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>
@@ -62,8 +74,8 @@ hide:
           </td>
           <td>2024-06-04</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>5</td>
+          <td>1</td>
+          <td>6</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.">
@@ -86,7 +98,7 @@ hide:
           </td>
           <td>2023-10-12</td>
           <td>ArXiv</td>
-          <td>23</td>
+          <td>29</td>
           <td>40</td>
         </tr>
     
@@ -98,7 +110,7 @@ hide:
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>69</td>
+          <td>82</td>
           <td>18</td>
         </tr>
     
@@ -110,8 +122,8 @@ hide:
           </td>
           <td>2024-10-15</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>27</td>
+          <td>1</td>
+          <td>28</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).">
@@ -122,22 +134,10 @@ hide:
           </td>
           <td>2024-05-03</td>
           <td>ArXiv</td>
-          <td>7</td>
+          <td>13</td>
           <td>47</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.">
-          <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>
-            Yuxuan Bian, Xu Ju, Jiangtong Li, Zhijian Xu, Dawei Cheng, Qiang Xu
-          </td>
-          <td>2024-02-07</td>
-          <td>ArXiv</td>
-          <td>8</td>
-          <td>8</td>
-        </tr>
-    
   </tbody>
   <tfoot>
     <tr>
diff --git a/docs/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8.md b/docs/recommendations/e6f0a85009481dcfd93aaa43ed3f980e5033b0d8.md
index 72ee5f26..586bf237 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 2024-11-11 06:05:53 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:30:19 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -62,8 +62,8 @@ hide:
           </td>
           <td>2015-09-11</td>
           <td>Proceedings of the National Academy of Sciences</td>
-          <td>3408</td>
-          <td>67</td>
+          <td>3467</td>
+          <td>68</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.">
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2023-10-06</td>
           <td>2024 American Control Conference (ACC)</td>
-          <td>1</td>
+          <td>2</td>
           <td>1</td>
         </tr>
     
@@ -86,8 +86,8 @@ hide:
           </td>
           <td>2021-11-12</td>
           <td>J. Open Source Softw.</td>
-          <td>119</td>
-          <td>67</td>
+          <td>124</td>
+          <td>68</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.">
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2020-05-05</td>
           <td>Nature Communications</td>
-          <td>289</td>
+          <td>300</td>
           <td>13</td>
         </tr>
     
@@ -122,8 +122,8 @@ hide:
           </td>
           <td>2020-04-17</td>
           <td>arXiv: Dynamical Systems</td>
-          <td>51</td>
-          <td>67</td>
+          <td>52</td>
+          <td>68</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.">
@@ -138,6 +138,18 @@ hide:
           <td>24</td>
         </tr>
     
+        <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>56</td>
+          <td>68</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.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
           <td><a href="https://www.semanticscholar.org/paper/8787d66296c0337aeb54d87885d29a63af17711b" target='_blank'>Sparse learning of stochastic dynamical equations.</a></td>
@@ -146,7 +158,7 @@ hide:
           </td>
           <td>2017-12-06</td>
           <td>The Journal of chemical physics</td>
-          <td>200</td>
+          <td>202</td>
           <td>46</td>
         </tr>
     
diff --git a/docs/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f.md b/docs/recommendations/f45f85fa1beaa795c24c4ff86f1f2deece72252f.md
index 0fe04551..7c2b204f 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 2024-11-11 06:05:15 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:36 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2023-10-12</td>
           <td>ArXiv</td>
-          <td>23</td>
+          <td>29</td>
           <td>40</td>
         </tr>
     
@@ -74,7 +74,7 @@ hide:
           </td>
           <td>2024-06-22</td>
           <td>ArXiv</td>
-          <td>13</td>
+          <td>18</td>
           <td>3</td>
         </tr>
     
@@ -98,7 +98,7 @@ hide:
           </td>
           <td>2023-10-03</td>
           <td>ArXiv</td>
-          <td>195</td>
+          <td>213</td>
           <td>9</td>
         </tr>
     
@@ -110,7 +110,7 @@ hide:
           </td>
           <td>2024-05-23</td>
           <td>ArXiv</td>
-          <td>2</td>
+          <td>3</td>
           <td>2</td>
         </tr>
     
@@ -122,32 +122,32 @@ hide:
           </td>
           <td>2024-10-15</td>
           <td>ArXiv</td>
-          <td>0</td>
-          <td>27</td>
+          <td>1</td>
+          <td>28</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 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/663dc434fabc1cf1d3e85fff3f7ddcd313035d18" target='_blank'>LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/30abc3fad3c88886c189e216c8414b18c396c6bd" target='_blank'>A Mamba Foundation Model for Time Series Forecasting</a></td>
           <td>
-            Ching Chang, Wenjie Peng, Tien-Fu Chen
+            Haoyu Ma, Yushu Chen, Wenlai Zhao, Jinzhe Yang, Yingsheng Ji, Xinghua Xu, Xiaozhu Liu, Hao Jing, Shengzhuo Liu, Guangwen Yang
           </td>
-          <td>2023-08-16</td>
+          <td>2024-11-05</td>
           <td>ArXiv</td>
-          <td>22</td>
-          <td>3</td>
+          <td>0</td>
+          <td>8</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.">
+        <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/2cb3044ef42c7ee022a988864028b80ce977072c" target='_blank'>Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts</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>
-            X. Shi, Shiyu Wang, Yuqi Nie, Dianqi Li, Zhou Ye, Qingsong Wen, Ming Jin
+            Ching Chang, Wenjie Peng, Tien-Fu Chen
           </td>
-          <td>2024-09-24</td>
+          <td>2023-08-16</td>
           <td>ArXiv</td>
-          <td>2</td>
-          <td>9</td>
+          <td>24</td>
+          <td>3</td>
         </tr>
     
   </tbody>
diff --git a/docs/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90.md b/docs/recommendations/ff1f1cf9df8c413ec7345da7604ba28597da5b90.md
index 6d974095..d8c65dd0 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 2024-11-11 06:05:30 UTC</i>
+  <i class="footer">This page was last updated on 2024-12-25 11:29:51 UTC</i>
   </p>
   
   <div class="note info" onclick="startIntro()">
@@ -26,7 +26,7 @@ hide:
 
   <p>
   <h3 data-intro='Recommendations for the article'>
-    Recommendations for the article <i>UNITS: A Unified Multi-Task Time Series Model</i>
+    Recommendations for the article <i>UniTS: A Unified Multi-Task Time Series Model</i>
   </h3>
   <table id="table1" class="display wrap" style="width:100%">
   <thead>
@@ -50,7 +50,7 @@ hide:
           </td>
           <td>2024-02-04</td>
           <td>ArXiv, DBLP</td>
-          <td>21</td>
+          <td>26</td>
           <td>67</td>
         </tr>
     
@@ -58,7 +58,7 @@ hide:
           <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
+            Özgün Turgut, Philip Müller, M. Menten, Daniel Rueckert
           </td>
           <td>2024-10-09</td>
           <td>ArXiv</td>
@@ -66,16 +66,16 @@ hide:
           <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.">
+        <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 time series. Technically, we propose a universal TimeAttention to facilitate generative Transformers on multiple time series, which can effectively capture fine-grained intra- and inter-series dependencies of flattened time series tokens (patches), and is further enhanced by deftly designed position embeddings for the temporal and variable dimensions. Timer-XL achieves state-of-the-art performance across challenging forecasting benchmarks through a unified approach. Based on large-scale pre-training, Timer-XL also demonstrates notable zero-shot performance, making it a promising architecture for large time series models.">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/d8140a17dfc21b1e3300158cb0676600136f14b1" target='_blank'>Large Pre-trained time series models for cross-domain Time series analysis tasks</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/518ecc43f230938e62b1f237ce9a0ef0d1f6d449" target='_blank'>Timer-XL: Long-Context Transformers for Unified Time Series Forecasting</a></td>
           <td>
-            Harshavardhan Kamarthi, B. A. Prakash
+            Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long
           </td>
-          <td>2023-11-19</td>
+          <td>2024-10-07</td>
           <td>ArXiv</td>
-          <td>4</td>
-          <td>8</td>
+          <td>0</td>
+          <td>67</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.">
@@ -102,18 +102,6 @@ hide:
           <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.">
-          <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>
-            Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin
-          </td>
-          <td>2023-11-24</td>
-          <td>ArXiv</td>
-          <td>4</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/02fa77e4f355198cb4270f6d4a07517bf09c46dd" target='_blank'>Chronos: Learning the Language of Time Series</a></td>
@@ -122,20 +110,20 @@ hide:
           </td>
           <td>2024-03-12</td>
           <td>ArXiv</td>
-          <td>69</td>
+          <td>82</td>
           <td>18</td>
         </tr>
     
-        <tr id="Deep learning for time series forecasting has traditionally operated within a one-model-per-dataset framework, limiting its potential to leverage the game-changing impact of large pre-trained models. The concept of universal forecasting, emerging from pre-training on a vast collection of time series datasets, envisions a single Large Time Series Model capable of addressing diverse downstream forecasting tasks. However, constructing such a model poses unique challenges specific to time series data: i) cross-frequency learning, ii) accommodating an arbitrary number of variates for multivariate time series, and iii) addressing the varying distributional properties inherent in large-scale data. To address these challenges, we present novel enhancements to the conventional time series Transformer architecture, resulting in our proposed Masked Encoder-based Universal Time Series Forecasting Transformer (Moirai). Trained on our newly introduced Large-scale Open Time Series Archive (LOTSA) featuring over 27B observations across nine domains, Moirai achieves competitive or superior performance as a zero-shot forecaster when compared to full-shot models. Code, data, and model weights can be found at https://github.com/SalesforceAIResearch/uni2ts.">
+        <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 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. Code: www.github.com/AdityaLab/Samay">
           <td id="tag"><i class="material-icons">visibility_off</i></td>
-          <td><a href="https://www.semanticscholar.org/paper/4a111f7a3b56d0468f13104999844885157ef17d" target='_blank'>Unified Training of Universal Time Series Forecasting Transformers</a></td>
+          <td><a href="https://www.semanticscholar.org/paper/d8140a17dfc21b1e3300158cb0676600136f14b1" target='_blank'>Large Pre-trained time series models for cross-domain Time series analysis tasks</a></td>
           <td>
-            Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, Doyen Sahoo
+            Harshavardhan Kamarthi, B. A. Prakash
           </td>
-          <td>2024-02-04</td>
+          <td>2023-11-19</td>
           <td>ArXiv</td>
-          <td>71</td>
-          <td>25</td>
+          <td>4</td>
+          <td>9</td>
         </tr>
     
   </tbody>

2024-02-04	ArXiv, DBLP	21	26	67
	Towards Generalisable Time Series Understanding Across Domains	- Ozgun Turgut, Philip Muller, M. Menten, Daniel Rueckert + Özgün Turgut, Philip Müller, M. Menten, Daniel Rueckert	2024-10-09	ArXiv	15
	Large Pre-trained time series models for cross-domain Time series analysis tasks	Timer-XL: Long-Context Transformers for Unified Time Series Forecasting	- Harshavardhan Kamarthi, B. A. Prakash + Yong Liu, Guo Qin, Xiangdong Huang, Jianmin Wang, Mingsheng Long	2023-11-19	2024-10-07	ArXiv	4	8	0	67
7
	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
	Chronos: Learning the Language of Time Series	2024-03-12	ArXiv	69	82	18
	Unified Training of Universal Time Series Forecasting Transformers	Large Pre-trained time series models for cross-domain Time series analysis tasks	- Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, Doyen Sahoo + Harshavardhan Kamarthi, B. A. Prakash	2024-02-04	2023-11-19	ArXiv	71	25	4	9