2020.05.15.txt

==========New Papers==========
1, TITLE:       Rolling Horizon NEAT for General Video Game Playing
http://arxiv.org/abs/2005.06764
AUTHORS:        Diego Perez-Liebana ; Muhammad Sajid Alam ; Raluca D. Gaina
COMMENTS:       8 pages, 5 figures, accepted for publication in IEEE Conference on Games (CoG) 2020
HIGHLIGHT:      This paper presents a new Statistical Forward Planning (SFP) method, Rolling Horizon NeuroEvolution of Augmenting Topologies (rhNEAT).

2, TITLE:       Towards Automatic building of Human-Machine Conversational System to support Maintenance Processes
http://arxiv.org/abs/2005.06517
AUTHORS:        Elena Coli ; Nicola Melluso ; Gualtiero Fantoni ; Daniele Mazzei
HIGHLIGHT:      So, the goal of this research is designing a methodology for automatic building of human-machine conversational system, able to interact in an industrial environment.

3, TITLE:       A Survey on Temporal Reasoning for Temporal Information Extraction from Text (Extended Abstract)
http://arxiv.org/abs/2005.06527
AUTHORS:        Artuur Leeuwenberg ; Marie-Francine Moens
COMMENTS:       Extended abstract of a JAIR article to appear in the proceedings of IJCAI 2020
HIGHLIGHT:      This article presents a comprehensive survey of the research from the past decades on temporal reasoning for automatic temporal information extraction from text, providing a case study on the integration of symbolic reasoning with machine learning-based information extraction systems.

4, TITLE:       Deep Learning for Political Science
http://arxiv.org/abs/2005.06540
AUTHORS:        Kakia Chatsiou ; Slava Jankin Mikhaylov
HIGHLIGHT:      Political science, and social science in general, have traditionally been using computational methods to study areas such as voting behavior, policy making, international conflict, and international development.

5, TITLE:       Validation and Normalization of DCS corpus using Sanskrit Heritage tools to build a tagged Gold Corpus
http://arxiv.org/abs/2005.06545
AUTHORS:        Sriram Krishnan ; Amba Kulkarni ; Gérard Huet
HIGHLIGHT:      This paper describes the modified alignment process in detail and records the additional linguistic differences observed.

6, TITLE:       Robust Visual Object Tracking with Two-Stream Residual Convolutional Networks
http://arxiv.org/abs/2005.06536
AUTHORS:        Ning Zhang ; Jingen Liu ; Ke Wang ; Dan Zeng ; Tao Mei
HIGHLIGHT:      Inspired by the human "visual tracking" capability which leverages motion cues to distinguish the target from the background, we propose a Two-Stream Residual Convolutional Network (TS-RCN) for visual tracking, which successfully exploits both appearance and motion features for model update.

7, TITLE:       A Mixture of $h-1$ Heads is Better than $h$ Heads
http://arxiv.org/abs/2005.06537
AUTHORS:        Hao Peng ; Roy Schwartz ; Dianqi Li ; Noah A. Smith
COMMENTS:       ACL2020
HIGHLIGHT:      In this work, we instead "reallocate" them -- the model learns to activate different heads on different inputs.

8, TITLE:       Detector-SegMentor Network for Skin Lesion Localization and Segmentation
http://arxiv.org/abs/2005.06550
AUTHORS:        Shreshth Saini ; Divij Gupta ; Anil Kumar Tiwari
COMMENTS:       9 pages, 7 figures, accepted at NCVPRIPG 2019
HIGHLIGHT:      In this study, we propose a simple yet novel network-in-network convolution neural network(CNN) based approach for segmentation of the skin lesion.

9, TITLE:       Arabic Dialect Identification in the Wild
http://arxiv.org/abs/2005.06557
AUTHORS:        Ahmed Abdelali ; Hamdy Mubarak ; Younes Samih ; Sabit Hassan ; Kareem Darwish
COMMENTS:       13 pages, 7 figures, 4 tables
HIGHLIGHT:      We present QADI, an automatically collected dataset of tweets belonging to a wide range of country-level Arabic dialects -covering 18 different countries in the MENA (Middle East and North Africa) region.

10, TITLE:      Document-Level Event Role Filler Extraction using Multi-Granularity Contextualized Encoding
http://arxiv.org/abs/2005.06579
AUTHORS:        Xinya Du ; Claire Cardie
COMMENTS:       Accepted to ACL 2020 (long papers), 12 pages
HIGHLIGHT:      To dynamically aggregate information captured by neural representations learned at different levels of granularity (e.g., the sentence- and paragraph-level), we propose a novel multi-granularity reader.

11, TITLE:      Pedestrian Action Anticipation using Contextual Feature Fusion in Stacked RNNs
http://arxiv.org/abs/2005.06582
AUTHORS:        Amir Rasouli ; Iuliia Kotseruba ; John K. Tsotsos
COMMENTS:       This paper was accepted and presented at British Machine Vision Conference (BMVC) 2019
HIGHLIGHT:      To this end, we propose a solution for the problem of pedestrian action anticipation at the point of crossing.

12, TITLE:      Do Saliency Models Detect Odd-One-Out Targets? New Datasets and Evaluations
http://arxiv.org/abs/2005.06583
AUTHORS:        Iuliia Kotseruba ; Calden Wloka ; Amir Rasouli ; John K. Tsotsos
COMMENTS:       Published in BMVC 2019. 14 pages, 5 figures
HIGHLIGHT:      We introduce two novel datasets, one with psychophysical patterns and one with natural odd-one-out stimuli.

13, TITLE:      Entity-Enriched Neural Models for Clinical Question Answering
http://arxiv.org/abs/2005.06587
AUTHORS:        Bhanu Pratap Singh Rawat ; Wei-Hung Weng ; Preethi Raghavan ; Peter Szolovits
HIGHLIGHT:      We explore state-of-the-art neural models for question answering on electronic medical records and improve their ability to generalize better on previously unseen (paraphrased) questions at test time.

14, TITLE:      PERLEX: A Bilingual Persian-English Gold Dataset for Relation Extraction
http://arxiv.org/abs/2005.06588
AUTHORS:        Majid Asgari-Bidhendi ; Mehrdad Nasser ; Behrooz Janfada ; Behrouz Minaei-Bidgoli
HIGHLIGHT:      In this paper, we present "PERLEX" as the first Persian dataset for relation extraction, which is an expert-translated version of the "Semeval-2010-Task-8" dataset.

15, TITLE:      Formal Analysis and Redesign of a Neural Network-Based Aircraft Taxiing System with VerifAI
http://arxiv.org/abs/2005.07173
AUTHORS:        Daniel J. Fremont ; Johnathan Chiu ; Dragos D. Margineantu ; Denis Osipychev ; Sanjit A. Seshia
COMMENTS:       Full version of a CAV 2020 paper
HIGHLIGHT:      We demonstrate a unified approach to rigorous design of safety-critical autonomous systems using the VerifAI toolkit for formal analysis of AI-based systems.

16, TITLE:      Estimating predictive uncertainty for rumour verification models
http://arxiv.org/abs/2005.07174
AUTHORS:        Elena Kochkina ; Maria Liakata
COMMENTS:       Accepted to the Annual Conference of the Association for Computational Linguistics (ACL) 2020
HIGHLIGHT:      We propose two methods for uncertainty-based instance rejection, supervised and unsupervised.

17, TITLE:      OctSqueeze: Octree-Structured Entropy Model for LiDAR Compression
http://arxiv.org/abs/2005.07178
AUTHORS:        Lila Huang ; Shenlong Wang ; Kelvin Wong ; Jerry Liu ; Raquel Urtasun
COMMENTS:       CVPR 2020 (Oral)
HIGHLIGHT:      We present a novel deep compression algorithm to reduce the memory footprint of LiDAR point clouds.

18, TITLE:      DRTS Parsing with Structure-Aware Encoding and Decoding
http://arxiv.org/abs/2005.06901
AUTHORS:        Qiankun Fu ; Yue Zhang ; Jiangming Liu ; Meishan Zhang
COMMENTS:       ACL2020
HIGHLIGHT:      In this work, we propose a structural-aware model at both the encoder and decoder phase to integrate the structural information, where graph attention network (GAT) is exploited for effectively modeling.

19, TITLE:      Classification of Arrhythmia by Using Deep Learning with 2-D ECG Spectral Image Representation
http://arxiv.org/abs/2005.06902
AUTHORS:        Amin Ullah ; Syed M. Anwar ; Muhammad Bilal ; Raja M Mehmood
COMMENTS:       14 pages, 5 figures, accepted for future publication in Remote Sensing MDPI Journal
HIGHLIGHT:      In this study, we propose a two-dimensional (2-D) convolutional neural network (CNN) model for the classification of ECG signals into eight classes; namely, normal beat, premature ventricular contraction beat, paced beat, right bundle branch block beat, left bundle branch block beat, atrial premature contraction beat, ventricular flutter wave beat, and ventricular escape beat.

20, TITLE:      Cognitive Amplifier for Internet of Things
http://arxiv.org/abs/2005.06914
AUTHORS:        Bing Huang ; Athman Bouguettaya ; Azadeh Ghari Neiat
HIGHLIGHT:      We present a Cognitive Amplifier framework to augment things part of an IoT, with cognitive capabilities for the purpose of improving life convenience.

21, TITLE:      Manthan: A Data Driven Approach for Boolean Function Synthesis
http://arxiv.org/abs/2005.06922
AUTHORS:        Priyanka Golia ; Subhajit Roy ; Kuldeep S. Meel
COMMENTS:       24 pages including references, and 8 figures. To be published in 32nd International Conference on Computer-Aided Verification (CAV-2020)
HIGHLIGHT:      Motivated by the progress in machine learning, we propose Manthan, a novel data-driven approach to Boolean functional synthesis.

22, TITLE:      NIT-Agartala-NLP-Team at SemEval-2020 Task 8: Building Multimodal Classifiers to tackle Internet Humor
http://arxiv.org/abs/2005.06943
AUTHORS:        Steve Durairaj Swamy ; Shubham Laddha ; Basil Abdussalam ; Debayan Datta ; Anupam Jamatia
HIGHLIGHT:      The paper describes the systems submitted to SemEval-2020 Task 8: Memotion by the `NIT-Agartala-NLP-Team'.

23, TITLE:      Noise Homogenization via Multi-Channel Wavelet Filtering for High-Fidelity Sample Generation in GANs
http://arxiv.org/abs/2005.06707
AUTHORS:        Shaoning Zeng ; Bob Zhang
COMMENTS:       12 pages, 2 figures
HIGHLIGHT:      In this work, we propose a novel multi-channel wavelet-based filtering method for GANs, to cope with this problem.

24, TITLE:      Domain Conditioned Adaptation Network
http://arxiv.org/abs/2005.06717
AUTHORS:        Shuang Li ; Chi Harold Liu ; Qiuxia Lin ; Binhui Xie ; Zhengming Ding ; Gao Huang ; Jian Tang
COMMENTS:       Accepted by AAAI 2020
HIGHLIGHT:      In this paper, we relax a shared-convnets assumption made by previous DA methods and propose a Domain Conditioned Adaptation Network (DCAN), which aims to excite distinct convolutional channels with a domain conditioned channel attention mechanism.

25, TITLE:      Enhanced Residual Networks for Context-based Image Outpainting
http://arxiv.org/abs/2005.06723
AUTHORS:        Przemek Gardias ; Eric Arthur ; Huaming Sun
COMMENTS:       6 pages, 5 figures
HIGHLIGHT:      We propose two methods to improve this issue: the use of a local and global discriminator, and the addition of residual blocks within the encoding section of the network.

26, TITLE:      Low-Dose CT Image Denoising Using Parallel-Clone Networks
http://arxiv.org/abs/2005.06724
AUTHORS:        Siqi Li ; Guobao Wang
HIGHLIGHT:      In this paper, we propose a parallel-clone neural network method that utilizes a modularized network model and exploits the benefit of parallel input, parallel-output loss, and clone-toclone feature transfer.

27, TITLE:      S2IGAN: Speech-to-Image Generation via Adversarial Learning
http://arxiv.org/abs/2005.06968
AUTHORS:        Xinsheng Wang ; Tingting Qiao ; Jihua Zhu ; Alan Hanjalic ; Odette Scharenborg
HIGHLIGHT:      In this paper, a speech-to-image generation (S2IG) framework is proposed which translates speech descriptions to photo-realistic images without using any text information, thus allowing unwritten languages to potentially benefit from this technology.

28, TITLE:      OD-SGD: One-step Delay Stochastic Gradient Descent for Distributed Training
http://arxiv.org/abs/2005.06728
AUTHORS:        Yemao Xu ; Dezun Dong ; Weixia Xu ; Xiangke Liao
HIGHLIGHT:      To sufficiently utilize the advantages of SSGD and ASGD, we propose a novel technology named One-step Delay SGD (OD-SGD) to combine their strengths in the training process.

29, TITLE:      Dense-Resolution Network for Point Cloud Classification and Segmentation
http://arxiv.org/abs/2005.06734
AUTHORS:        Shi Qiu ; Saeed Anwar ; Nick Barnes
HIGHLIGHT:      In this article, we propose a novel network named Dense-Resolution Network for point cloud analysis.

30, TITLE:      The Information & Mutual Information Ratio for Counting Image Features and Their Matches
http://arxiv.org/abs/2005.06739
AUTHORS:        Ali Khajegili Mirabadi ; Stefano Rini
COMMENTS:       8-th Iran Workshop on Communication and Information Theory, 2020
HIGHLIGHT:      In this paper, two new image features are proposed: the Information Ratio (IR) and the Mutual Information Ratio (MIR).

31, TITLE:      Large Scale Font Independent Urdu Text Recognition System
http://arxiv.org/abs/2005.06752
AUTHORS:        Atique Ur Rehman ; Sibt Ul Hussain
HIGHLIGHT:      Urdu is among the languages which did not receive much attention, especially in the font independent perspective.

32, TITLE:      Many-Objective Software Remodularization using NSGA-III
http://arxiv.org/abs/2005.06510
AUTHORS:        Mohamed Wiem Mkaouer ; Marouane Kessentini ; Adnan Shaout ; Patrice Koligheu ; Slim Bechikh ; Kalyanmoy Deb ; Ali Ouni
COMMENTS:       Mkaouer, Wiem, et al. "Many-objective software remodularization using NSGA-III." ACM Transactions on Software Engineering and Methodology (TOSEM) 24.3 (2015): 1-45
HIGHLIGHT:      In this paper, we propose a novel many-objective search-based approach using NSGA-III.

33, TITLE:      Generative Models for Generic Light Field Reconstruction
http://arxiv.org/abs/2005.06508
AUTHORS:        Paramanand Chandramouli ; Kanchana Vaishnavi Gandikota ; Andreas Goerlitz ; Andreas Kolb ; Michael Moeller
HIGHLIGHT:      In this work, we present for the first time generative models for 4D light field patches using variational autoencoders to capture the data distribution of light field patches.

34, TITLE:      A Rate-Distortion view of human pragmatic reasoning
http://arxiv.org/abs/2005.06641
AUTHORS:        Noga Zaslavsky ; Jennifer Hu ; Roger P. Levy
HIGHLIGHT:      Here, we present a novel analysis of the RSA framework that addresses this question.

35, TITLE:      Enabling Edge Cloud Intelligence for Activity Learning in Smart Home
http://arxiv.org/abs/2005.06885
AUTHORS:        Bing Huang ; Athman Bouguettaya ; Hai Dong
HIGHLIGHT:      In this paper, we utilize temporal features for activity recognition and prediction in a single smart home setting.

36, TITLE:      A Generating-Extension-Generator for Machine Code
http://arxiv.org/abs/2005.06645
AUTHORS:        Michael Vaughn ; Thomas Reps
COMMENTS:       21 pages, 8 Figures
HIGHLIGHT:      This paper presents a new approach to state management in a program specializer.

37, TITLE:      ZynqNet: An FPGA-Accelerated Embedded Convolutional Neural Network
http://arxiv.org/abs/2005.06892
AUTHORS:        David Gschwend
COMMENTS:       85 pages, 26 figures. Code available at http://github.com/dgschwend/zynqnet
HIGHLIGHT:      This master thesis explores the potential of FPGA-based CNN acceleration and demonstrates a fully functional proof-of-concept CNN implementation on a Zynq System-on-Chip.

38, TITLE:      Conflict Detection of IoT_Services in Smart Home
http://arxiv.org/abs/2005.06895
AUTHORS:        Bing Huang ; Athman Bouguettaya ; Sajib Mistry
HIGHLIGHT:      We propose a novel framework that detects conflicts in IoT-based smart homes.

39, TITLE:      Structured Query-Based Image Retrieval Using Scene Graphs
http://arxiv.org/abs/2005.06653
AUTHORS:        Brigit Schroeder ; Subarna Tripathi
COMMENTS:       Accepted to Diagram Image Retrieval and Analysis (DIRA) Workshop at CVPR 2020
HIGHLIGHT:      In this paper we present a method which uses scene graph embeddings as the basis for an approach to image retrieval.

40, TITLE:      Flexible Example-based Image Enhancement with Task Adaptive Global Feature Self-Guided Network
http://arxiv.org/abs/2005.06654
AUTHORS:        Dario Kneubuehler ; Shuhang Gu ; Luc Van Gool ; Radu Timofte
HIGHLIGHT:      We propose the first practical multitask image enhancement network, that is able to learn one-to-many and many-to-one image mappings.

41, TITLE:      Mitigating Gender Bias in Machine Learning Data Sets
http://arxiv.org/abs/2005.06898
AUTHORS:        Susan Leavy ; Gerardine Meaney ; Karen Wade ; Derek Greene
COMMENTS:       10 pages, 5 figures, 5 Tables, Presented as Bias2020 workshop (as part of the ECIR Conference) - http://bias.disim.univaq.it
HIGHLIGHT:      This paper proposes a framework for the identification of gender bias in training data for machine learning.The work draws upon gender theory and sociolinguistics to systematically indicate levels of bias in textual training data and associated neural word embedding models, thus highlighting pathways for both removing bias from training data and critically assessing its impact in the context of search and recommender systems.

42, TITLE:      Exploiting Multi-Layer Grid Maps for Surround-View Semantic Segmentation of Sparse LiDAR Data
http://arxiv.org/abs/2005.06667
AUTHORS:        Frank Bieder ; Sascha Wirges ; Johannes Janosovits ; Sven Richter ; Zheyuan Wang ; Christoph Stiller
HIGHLIGHT:      In this paper, we consider the transformation of laser range measurements into a top-view grid map representation to approach the task of LiDAR-only semantic segmentation.

43, TITLE:      The Extended Theory of Trees and Algebraic (Co)datatypes
http://arxiv.org/abs/2005.06659
AUTHORS:        Fabian Zaiser ; C. -H. Luke Ong
COMMENTS:       full version of a paper submitted to HCVS 2020
HIGHLIGHT:      Based on their work, we present a simplification procedure that determines whether any given (not necessarily closed) formula is satisfiable, returning a simplified formula which enables one to read off all possible models.

44, TITLE:      Explaining Black Box Predictions and Unveiling Data Artifacts through Influence Functions
http://arxiv.org/abs/2005.06676
AUTHORS:        Xiaochuang Han ; Byron C. Wallace ; Yulia Tsvetkov
COMMENTS:       ACL 2020
HIGHLIGHT:      In this work, we investigate the use of influence functions for NLP, providing an alternative approach to interpreting neural text classifiers.

45, TITLE:      Activation functions are not needed: the ratio net
http://arxiv.org/abs/2005.06678
AUTHORS:        Chi-Chun Zhou ; Hai-Long Tu ; Yi Liu ; Jian Hua
HIGHLIGHT:      In this paper, we propose a new network that is efficient in finding the function that maps the feature to the label.

46, TITLE:      W-Cell-Net: Multi-frame Interpolation of Cellular Microscopy Videos
http://arxiv.org/abs/2005.06684
AUTHORS:        Rohit Saha ; Abenezer Teklemariam ; Ian Hsu ; Alan M. Moses
HIGHLIGHT:      We propose a fully convolutional autoencoder network that takes as input two images and generates upto seven intermediate images.

47, TITLE:      Symbolic Partial-Order Execution for Testing Multi-Threaded Programs
http://arxiv.org/abs/2005.06688
AUTHORS:        Daniel Schemmel ; Julian Büning ; César Rodríguez ; David Laprell ; Klaus Wehrle
COMMENTS:       Extended version of a paper accepted for publication in CAV'20
HIGHLIGHT:      We describe a technique for systematic testing of multi-threaded programs.

48, TITLE:      Surrogate Assisted Optimisation for Travelling Thief Problems
http://arxiv.org/abs/2005.06695
AUTHORS:        Majid Namazi ; Conrad Sanderson ; M. A. Hakim Newton ; Abdul Sattar
HIGHLIGHT:      We propose to make the search more efficient through an adaptive surrogate model (based on a customised form of Support Vector Regression) that learns the characteristics of initial TSP tours that lead to good TTP solutions.

49, TITLE:      FlowCFL: A Framework for Type-based Reachability Analysis in the Presence of Mutable Data
http://arxiv.org/abs/2005.06496
AUTHORS:        Ana Milanova
HIGHLIGHT:      We present FlowCFL, a framework for type-based reachability analysis in the presence of mutable data.

50, TITLE:      ZeroShotCeres: Zero-Shot Relation Extraction from Semi-Structured Webpages
http://arxiv.org/abs/2005.07105
AUTHORS:        Colin Lockard ; Prashant Shiralkar ; Xin Luna Dong ; Hannaneh Hajishirzi
COMMENTS:       Accepted to ACL 2020
HIGHLIGHT:      In this work, we propose a solution for "zero-shot" open-domain relation extraction from webpages with a previously unseen template, including from websites with little overlap with existing sources of knowledge for distant supervision and websites in entirely new subject verticals.

51, TITLE:      Robust On-Manifold Optimization for Uncooperative Space Relative Navigation with a Single Camera
http://arxiv.org/abs/2005.07110
AUTHORS:        Duarte Rondao ; Nabil Aouf ; Mark A. Richardson ; Vincent Dubanchet
COMMENTS:       42 pages, 17 figures
HIGHLIGHT:      In this paper, an innovative model-based approach is instead demonstrated to estimate the six-dimensional pose of a target object relative to the chaser spacecraft using solely a monocular setup.

52, TITLE:      Distilling neural networks into skipgram-level decision lists
http://arxiv.org/abs/2005.07111
AUTHORS:        Madhumita Sushil ; Simon Šuster ; Walter Daelamans
HIGHLIGHT:      To overcome these limitations, we propose a pipeline to explain RNNs by means of decision lists (also called rules) over skipgrams. For evaluation of explanations, we create a synthetic sepsis-identification dataset, as well as apply our technique on additional clinical and sentiment analysis datasets.

53, TITLE:      PENNI: Pruned Kernel Sharing for Efficient CNN Inference
http://arxiv.org/abs/2005.07133
AUTHORS:        Shiyu Li ; Edward Hanson ; Hai Li ; Yiran Chen
HIGHLIGHT:      Based on this observation, we propose PENNI, a CNN model compression framework that is able to achieve model compactness and hardware efficiency simultaneously by (1) implementing kernel sharing in convolution layers via a small number of basis kernels and (2) alternately adjusting bases and coefficients with sparse constraints.

54, TITLE:      Named Entity Recognition as Dependency Parsing
http://arxiv.org/abs/2005.07150
AUTHORS:        Juntao Yu ; Bernd Bohnet ; Massimo Poesio
COMMENTS:       Accepted by ACL 2020
HIGHLIGHT:      In this paper, we use ideas from graph-based dependency parsing to provide our model a global view on the input via a biaffine model (Dozat and Manning, 2017).

55, TITLE:      Towards Understanding the Adversarial Vulnerability of Skeleton-based Action Recognition
http://arxiv.org/abs/2005.07151
AUTHORS:        Tianhang Zheng ; Sheng Liu ; Changyou Chen ; Junsong Yuan ; Baochun Li ; Kui Ren
HIGHLIGHT:      In this paper, we attempt to conduct a thorough study towards understanding the adversarial vulnerability of skeleton-based action recognition.

56, TITLE:      Competing in a Complex Hidden Role Game with Information Set Monte Carlo Tree Search
http://arxiv.org/abs/2005.07156
AUTHORS:        Jack Reinhardt
HIGHLIGHT:      In this paper, Single Observer Information Set Monte Carlo Tree Search (SO-ISMCTS) is applied to Secret Hitler, a popular social deduction board game that combines traditional hidden role mechanics with the randomness of a card deck.

57, TITLE:      You Do Not Need More Data: Improving End-To-End Speech Recognition by Text-To-Speech Data Augmentation
http://arxiv.org/abs/2005.07157
AUTHORS:        Aleksandr Laptev ; Roman Korostik ; Aleksey Svischev ; Andrei Andrusenko ; Ivan Medennikov ; Sergey Rybin
COMMENTS:       Submitted to Interspeech 2020
HIGHLIGHT:      We argue that, when the training data amount is low, this approach can allow an end-to-end model to reach hybrid systems' quality.

58, TITLE:      NAT: Noise-Aware Training for Robust Neural Sequence Labeling
http://arxiv.org/abs/2005.07162
AUTHORS:        Marcin Namysl ; Sven Behnke ; Joachim Köhler
COMMENTS:       Accepted to appear at ACL 2020
HIGHLIGHT:      To this end, we formulate the noisy sequence labeling problem, where the input may undergo an unknown noising process and propose two Noise-Aware Training (NAT) objectives that improve robustness of sequence labeling performed on perturbed input: Our data augmentation method trains a neural model using a mixture of clean and noisy samples, whereas our stability training algorithm encourages the model to create a noise-invariant latent representation.

59, TITLE:      On abstract F-systems. A graph-theoretic model for paradoxes involving a falsity predicate and its application to argumentation frameworks
http://arxiv.org/abs/2005.07050
AUTHORS:        Gustavo A. Bodanza
COMMENTS:       16 pages
HIGHLIGHT:      In this paper we present the F-systems model abstracting from all the features of the language in which the represented sentences are expressed.

60, TITLE:      Reinforced Coloring for End-to-End Instance Segmentation
http://arxiv.org/abs/2005.07058
AUTHORS:        Tuan Tran Anh ; Khoa Nguyen-Tuan ; Won-Ki Jeong
HIGHLIGHT:      To exploit the advantages of conventional single-object-per-step segmentation methods without impairing the scalability, we propose a novel iterative deep reinforcement learning agent that learns how to differentiate multiple objects in parallel.

61, TITLE:      Multi-agent Communication meets Natural Language: Synergies between Functional and Structural Language Learning
http://arxiv.org/abs/2005.07064
AUTHORS:        Angeliki Lazaridou ; Anna Potapenko ; Olivier Tieleman
COMMENTS:       to appear at ACL 2020
HIGHLIGHT:      We present a method for combining multi-agent communication and traditional data-driven approaches to natural language learning, with an end goal of teaching agents to communicate with humans in natural language.

62, TITLE:      On Learned Operator Correction
http://arxiv.org/abs/2005.07069
AUTHORS:        Sebastian Lunz ; Andreas Hauptmann ; Tanja Tarvainen ; Carola-Bibiane Schönlieb ; Simon Arridge
COMMENTS:       28 pages, 11 Figures
HIGHLIGHT:      This paper discusses the conceptual difficulty to learn such a forward model correction and proceeds to present a possible solution as forward-backward correction that explicitly corrects in both data and solution spaces.

63, TITLE:      Probabilistic Guarantees for Safe Deep Reinforcement Learning
http://arxiv.org/abs/2005.07073
AUTHORS:        Edoardo Bacci ; David Parker
HIGHLIGHT:      We propose MOSAIC, an algorithm for measuring the safety of deep reinforcement learning agents in stochastic settings.

64, TITLE:      FaceFilter: Audio-visual speech separation using still images
http://arxiv.org/abs/2005.07074
AUTHORS:        Soo-Whan Chung ; Soyeon Choe ; Joon Son Chung ; Hong-Goo Kang
COMMENTS:       Under submission as a conference paper. Video examples: https://youtu.be/ku9xoLh62E
HIGHLIGHT:      The objective of this paper is to separate a target speaker's speech from a mixture of two speakers using a deep audio-visual speech separation network.

65, TITLE:      Bayesian Bits: Unifying Quantization and Pruning
http://arxiv.org/abs/2005.07093
AUTHORS:        Mart van Baalen ; Christos Louizos ; Markus Nagel ; Rana Ali Amjad ; Ying Wang ; Tijmen Blankevoort ; Max Welling
HIGHLIGHT:      We introduce Bayesian Bits, a practical method for joint mixed precision quantization and pruning through gradient based optimization.

66, TITLE:      Ambient Sound Helps: Audiovisual Crowd Counting in Extreme Conditions
http://arxiv.org/abs/2005.07097
AUTHORS:        Di Hu ; Lichao Mou ; Qingzhong Wang ; Junyu Gao ; Yuansheng Hua ; Dejing Dou ; Xiao Xiang Zhu
HIGHLIGHT:      In this work, we introduce a novel task of audiovisual crowd counting, in which visual and auditory information are integrated for counting purposes. We collect a large-scale benchmark, named auDiovISual Crowd cOunting (DISCO) dataset, consisting of 1,935 images and the corresponding audio clips, and 170,270 annotated instances.

67, TITLE:      Stealthy and Efficient Adversarial Attacks against Deep Reinforcement Learning
http://arxiv.org/abs/2005.07099
AUTHORS:        Jianwen Sun ; Tianwei Zhang ; Xiaofei Xie ; Lei Ma ; Yan Zheng ; Kangjie Chen ; Yang Liu
HIGHLIGHT:      In this paper, we introduce two novel adversarial attack techniques to \emph{stealthily} and \emph{efficiently} attack the DRL agents.

68, TITLE:      TAM: Temporal Adaptive Module for Video Recognition
http://arxiv.org/abs/2005.06803
AUTHORS:        Zhaoyang Liu ; Limin Wang ; Wayne Wu ; Chen Qian ; Tong Lu
COMMENTS:       Technical report
HIGHLIGHT:      To effectively capture this diverse motion pattern, this paper presents a new temporal adaptive module (TAM) to generate video-specific kernels based on its own feature maps.

69, TITLE:      A Semi-Supervised Assessor of Neural Architectures
http://arxiv.org/abs/2005.06821
AUTHORS:        Yehui Tang ; Yunhe Wang ; Yixing Xu ; Hanting Chen ; Chunjing Xu ; Boxin Shi ; Chao Xu ; Qi Tian ; Chang Xu
HIGHLIGHT:      In contrast with classical performance predictor optimized in a fully supervised way, this paper suggests a semi-supervised assessor of neural architectures.

70, TITLE:      Detection and Retrieval of Out-of-Distribution Objects in Semantic Segmentation
http://arxiv.org/abs/2005.06831
AUTHORS:        Philipp Oberdiek ; Matthias Rottmann ; Gernot A. Fink
HIGHLIGHT:      We present a novel pipeline for semantic segmenation that detects out-of-distribution (OOD) segments by means of the deep neural network's prediction and performs image retrieval after feature extraction and dimensionality reduction on image patches.

71, TITLE:      Shortest Distances as Enumeration Problem
http://arxiv.org/abs/2005.06827
AUTHORS:        Katrin Casel ; Tobias Friedrich ; Stefan Neubert ; Markus L. Schmid
HIGHLIGHT:      We investigate the single source shortest distance (SSSD) and all pairs shortest distance (APSD) problems as enumeration problems (on unweighted and integer weighted graphs), meaning that the shortest distances are produced and listed one by one without repetition.

72, TITLE:      RegQCNET: Deep Quality Control for Image-to-template Brain MRI Registration
http://arxiv.org/abs/2005.06835
AUTHORS:        Baudouin Denis de Senneville ; José V. Manjon ; Pierrick Coupé
HIGHLIGHT:      In the current study, a compact 3D CNN, referred to as RegQCNET, is introduced to quantitatively predict the amplitude of a registration mismatch between the registered image and the reference template.

73, TITLE:      Industrial Federated Learning -- Requirements and System Design
http://arxiv.org/abs/2005.06850
AUTHORS:        Thomas Hiessl ; Daniel Schall ; Jana Kemnitz ; Stefan Schulte
COMMENTS:       12 pages, accepted for https://www.paams.net/workshops/agedai
HIGHLIGHT:      Therefore, we introduce an Industrial Federated Learning (IFL) system supporting knowledge exchange in continuously evaluated and updated FL cohorts of learning tasks with sufficient data similarity.

74, TITLE:      Ethical Adversaries: Towards Mitigating Unfairness with Adversarial Machine Learning
http://arxiv.org/abs/2005.06852
AUTHORS:        Pieter Delobelle ; Paul Temple ; Gilles Perrouin ; Benoît Frénay ; Patrick Heymans ; Bettina Berendt
COMMENTS:       17 pages, 3 figures, 1 table
HIGHLIGHT:      We offer a new framework that assists in mitigating unfair representations in the dataset used for training.

75, TITLE:      Towards Socially Responsible AI: Cognitive Bias-Aware Multi-Objective Learning
http://arxiv.org/abs/2005.06618
AUTHORS:        Procheta Sen ; Debasis Ganguly
HIGHLIGHT:      To alleviate this problem, we propose a bias-aware multi-objective learning framework that given a set of identity attributes (e.g. gender, ethnicity etc.) and a subset of sensitive categories of the possible classes of prediction outputs, learns to reduce the frequency of predicting certain combinations of them, e.g. predicting stereotypes such as `most blacks use abusive language', or `fear is a virtue of women'.

76, TITLE:      India nudges to contain COVID-19 pandemic: a reactive public policy analysis using machine-learning based topic modelling
http://arxiv.org/abs/2005.06619
AUTHORS:        Ramit Debnath ; Ronita Bardhan
COMMENTS:       35 pages with 4 figures, 4 boxes and 9 tables
HIGHLIGHT:      This study investigated how government formed reactive policies to fight coronavirus across its policy sectors.

77, TITLE:      A Category Theory Approach to Interoperability
http://arxiv.org/abs/2005.06872
AUTHORS:        Riccardo Del Gratta
COMMENTS:       Paper submitted to Applied Category Theory 2020
HIGHLIGHT:      In this article, we propose a Category Theory approach to (syntactic) interoperability between linguistic tools.



==========Updates to Previous Papers==========
1, TITLE:       Statistical Queries and Statistical Algorithms: Foundations and Applications
http://arxiv.org/abs/2004.00557
AUTHORS:        Lev Reyzin
COMMENTS:       21 pages
HIGHLIGHT:      We introduce the model, give the main definitions, and we explore the fundamental theory statistical queries and how how it connects to various notions of learnability.

2, TITLE:       A Survey on Deep Learning-based Architectures for Semantic Segmentation on 2D images
http://arxiv.org/abs/1912.10230
AUTHORS:        Irem Ulku ; Erdem Akagunduz
COMMENTS:       Updated with new studies
HIGHLIGHT:      In this survey, we mainly focus on the recent scientific developments in semantic segmentation, specifically on deep learning-based methods using 2D images.

3, TITLE:       SentiBERT: A Transferable Transformer-Based Architecture for Compositional Sentiment Semantics
http://arxiv.org/abs/2005.04114
AUTHORS:        Da Yin ; Tao Meng ; Kai-Wei Chang
COMMENTS:       ACL-2020
HIGHLIGHT:      We propose SentiBERT, a variant of BERT that effectively captures compositional sentiment semantics.

4, TITLE:       Data Augmentation Imbalance For Imbalanced Attribute Classification
http://arxiv.org/abs/2004.13628
AUTHORS:        Xiaying Bai ; Yang Hu ; Pan Zhou ; Fanhua Shang ; Shengmei Shen
COMMENTS:       This paper needs further revision
HIGHLIGHT:      In this paper, we propose a new re-sampling algorithm called: data augmentation imbalance (DAI) to explicitly enhance the ability to discriminate the fewer attributes via increasing the proportion of labels accounting for a small part.

5, TITLE:       Combining Deep Learning with Geometric Features for Image based Localization in the Gastrointestinal Tract
http://arxiv.org/abs/2005.05481
AUTHORS:        Jingwei Song ; Mitesh Patel ; Andreas Girgensohn ; Chelhwon Kim
HIGHLIGHT:      Considering these, we propose a novel approach to combine DL method with traditional feature based approach to achieve better localization with small training data.

6, TITLE:       Boson-Sampling in the light of sample complexity
http://arxiv.org/abs/1306.3995
AUTHORS:        C. Gogolin ; M. Kliesch ; L. Aolita ; J. Eisert
COMMENTS:       22 pages, 1 figure; v2: typos corrected and minor improvements; v3: unaltered. For an improved sampling complexity lower bound that, in particular, holds for general verification algorithms, see arXiv:1812.01023
HIGHLIGHT:      In this work we show that in this setup, with probability exponentially close to one in the number of bosons, no symmetric algorithm can distinguish the Boson-Sampling distribution from the uniform one from fewer than exponentially many samples.

7, TITLE:       Multi-modal Embedding Fusion-based Recommender
http://arxiv.org/abs/2005.06331
AUTHORS:        Anna Wroblewska ; Jacek Dabrowski ; Michal Pastuszak ; Andrzej Michalowski ; Michal Daniluk ; Barbara Rychalska ; Mikolaj Wieczorek ; Sylwia Sysko-Romanczuk
COMMENTS:       7 pages, 8 figures
HIGHLIGHT:      Here, we present our system, its flexibility and performance.

8, TITLE:       GLUECoS : An Evaluation Benchmark for Code-Switched NLP
http://arxiv.org/abs/2004.12376
AUTHORS:        Simran Khanuja ; Sandipan Dandapat ; Anirudh Srinivasan ; Sunayana Sitaram ; Monojit Choudhury
COMMENTS:       To appear at ACL 2020
HIGHLIGHT:      We present an evaluation benchmark, GLUECoS, for code-switched languages, that spans several NLP tasks in English-Hindi and English-Spanish.

9, TITLE:       RSVQA: Visual Question Answering for Remote Sensing Data
http://arxiv.org/abs/2003.07333
AUTHORS:        Sylvain Lobry ; Diego Marcos ; Jesse Murray ; Devis Tuia
COMMENTS:       12 pages, Published in IEEE Transactions on Geoscience and Remote Sensing. Added one experiment and authors' biographies
HIGHLIGHT:      This paper introduces the task of visual question answering for remote sensing data (RSVQA). Using an automatic method introduced in this article, we built two datasets (using low and high resolution data) of image/question/answer triplets.

10, TITLE:      RISE Video Dataset: Recognizing Industrial Smoke Emissions
http://arxiv.org/abs/2005.06111
AUTHORS:        Yen-Chia Hsu ; Ting-Hao 'Kenneth' Huang ; Ting-Yao Hu ; Paul Dille ; Sean Prendi ; Ryan Hoffman ; Anastasia Tsuhlares ; Randy Sargent ; Illah Nourbakhsh
COMMENTS:       Technical report
HIGHLIGHT:      We introduce RISE, the first large-scale video dataset for Recognizing Industrial Smoke Emissions.

11, TITLE:      Evidence Inference 2.0: More Data, Better Models
http://arxiv.org/abs/2005.04177
AUTHORS:        Jay DeYoung ; Eric Lehman ; Ben Nye ; Iain J. Marshall ; Byron C. Wallace
COMMENTS:       Accepted as workshop paper into BioNLP Updated results from SciBERT to Biomed RoBERTa
HIGHLIGHT:      In this paper, we collect additional annotations to expand the Evidence Inference dataset by 25\%, provide stronger baseline models, systematically inspect the errors that these make, and probe dataset quality.

12, TITLE:      Selective-Candidate Framework with Similarity Selection Rule for Evolutionary Optimization
http://arxiv.org/abs/1712.06338
AUTHORS:        Sheng Xin Zhang ; Wing Shing Chan ; Zi Kang Peng ; Shao Yong Zheng ; Kit Sang Tang
HIGHLIGHT:      To address these problems, this paper proposes an explicit EEC control method named selective-candidate framework with similarity selection rule (SCSS).

13, TITLE:      Improving Confidence Estimates for Unfamiliar Examples
http://arxiv.org/abs/1804.03166
AUTHORS:        Zhizhong Li ; Derek Hoiem
COMMENTS:       Accepted in CVPR 2020 (oral). ERRATA: a previous version (v3) included erroneous results for $T$-scaling, where novel samples are mistakenly included in the validation set for calibration. Please disregard the results of the previous version
HIGHLIGHT:      In this paper, we compare and evaluate several methods to improve confidence estimates for unfamiliar and familiar samples.

14, TITLE:      SRDA-Net: Super-Resolution Domain Adaptation Networks for Semantic Segmentation
http://arxiv.org/abs/2005.06382
AUTHORS:        Zhenjie Tang ; Bin Pan ; Enhai Liu ; Xia Xu ; Tianyang Shi ; Zhenwei Shi
HIGHLIGHT:      In this work, we design a novel end-to-end semantic segmentation network, Super- Resolution Domain Adaptation Network (SRDA-Net), which could simultaneously complete super-resolution and domain adaptation.

15, TITLE:      A Novel CNet-assisted Evolutionary Level Repairer and Its Applications to Super Mario Bros
http://arxiv.org/abs/2005.06148
AUTHORS:        Tianye Shu ; Ziqi Wang ; Jialin Liu ; Xin Yao
COMMENTS:       Accepted at IEEE CEC2020
HIGHLIGHT:      The proposed approaches are proved to be effective in our case study of repairing GAN-generated and artificially destroyed levels of Super Mario Bros. game.

16, TITLE:      SOLOIST: Few-shot Task-Oriented Dialog with A Single Pre-trained Auto-regressive Model
http://arxiv.org/abs/2005.05298
AUTHORS:        Baolin Peng ; Chunyuan Li ; Jinchao Li ; Shahin Shayandeh ; Lars Liden ; Jianfeng Gao
COMMENTS:       10 pages; Project Website: https://aka.ms/soloist
HIGHLIGHT:      This paper presents a new method SOLOIST, which uses transfer learning to efficiently build task-oriented dialog systems at scale.

17, TITLE:      Video Storytelling: Textual Summaries for Events
http://arxiv.org/abs/1807.09418
AUTHORS:        Junnan Li ; Yongkang Wong ; Qi Zhao ; Mohan S. Kankanhalli
COMMENTS:       Published in IEEE Transactions on Multimedia
HIGHLIGHT:      Bridging vision and natural language is a longstanding goal in computer vision and multimedia research. In this work, we introduce the problem of video storytelling, which aims at generating coherent and succinct stories for long videos.

18, TITLE:      A Survey on Knowledge Graph Embeddings with Literals: Which model links better Literal-ly?
http://arxiv.org/abs/1910.12507
AUTHORS:        Genet Asefa Gesese ; Russa Biswas ; Mehwish Alam ; Harald Sack
HIGHLIGHT:      This paper conducts a survey of KG embedding models which not only consider the structured information contained in the form of entities and relations in a KG but also the unstructured information represented as literals such as text, numerical values, images, etc.

19, TITLE:      Cross-Domain Complementary Learning Using Pose for Multi-Person Part Segmentation
http://arxiv.org/abs/1907.05193
AUTHORS:        Kevin Lin ; Lijuan Wang ; Kun Luo ; Yinpeng Chen ; Zicheng Liu ; Ming-Ting Sun
COMMENTS:       To appear in IEEE Transactions on Circuits and Systems for Video Technology; Presented at ICCV 2019 Demonstration
HIGHLIGHT:      In this paper, we observe that real and synthetic humans both have a skeleton (pose) representation.

20, TITLE:      Surrogate-assisted parallel tempering for Bayesian neural learning
http://arxiv.org/abs/1811.08687
AUTHORS:        Rohitash Chandra ; Konark Jain ; Arpit Kapoor ; Ashray Aman
COMMENTS:       Engineering Applications of Artificial Intelligence
HIGHLIGHT:      In this paper, we address the inefficiency of parallel tempering MCMC for large-scale problems by combining parallel computing features with surrogate assisted likelihood estimation that describes the plausibility of a model parameter value, given specific observed data.

21, TITLE:      A Deep Dive into Understanding Tumor Foci Classification using Multiparametric MRI Based on Convolutional Neural Network
http://arxiv.org/abs/1903.12331
AUTHORS:        Weiwei Zong ; Joon Lee ; Chang Liu ; Eric Carver ; Aharon Feldman ; Branislava Janic ; Mohamed Elshaikh ; Milan Pantelic ; David Hearshen ; Indrin Chetty ; Benjamin Movsas ; Ning Wen
HIGHLIGHT:      This work designs a customized workflow for the small and imbalanced data set of prostate mpMRI where features were extracted from a deep learning model and then analyzed by a traditional machine learning classifier.

22, TITLE:      An Assignment Problem Formulation for Dominance Move Indicator
http://arxiv.org/abs/2002.10842
AUTHORS:        Claudio Lucio do Val Lopes ; Flávio Vinícius Cruzeiro Martins ; Elizabeth F. Wanner
COMMENTS:       arXiv admin note: text overlap with arXiv:2001.03657
HIGHLIGHT:      The original formulation presents an efficient method to calculate it in a biobjective case only.

23, TITLE:      Relevance in Structured Argumentation
http://arxiv.org/abs/1809.04861
AUTHORS:        AnneMarie Borg ; Christian Straßer
COMMENTS:       Extended version of the paper with the same name published in the main track of IJCAI 2018. It countains additionally a treatment of credulous and weak skeptical semantics
HIGHLIGHT:      In this paper we investigate properties of structured argumentation systems that warrant relevance desiderata.

24, TITLE:      End-to-End Pseudo-LiDAR for Image-Based 3D Object Detection
http://arxiv.org/abs/2004.03080
AUTHORS:        Rui Qian ; Divyansh Garg ; Yan Wang ; Yurong You ; Serge Belongie ; Bharath Hariharan ; Mark Campbell ; Kilian Q. Weinberger ; Wei-Lun Chao
COMMENTS:       Accepted to 2020 Conference on Computer Vision and Pattern Recognition (CVPR 2020)
HIGHLIGHT:      In this paper, we introduce a new framework based on differentiable Change of Representation (CoR) modules that allow the entire PL pipeline to be trained end-to-end.

25, TITLE:      Homophone-based Label Smoothing in End-to-End Automatic Speech Recognition
http://arxiv.org/abs/2004.03437
AUTHORS:        Yi Zheng ; Xianjie Yang ; Xuyong Dang
HIGHLIGHT:      A new label smoothing method that makes use of prior knowledge of a language at human level, homophone, is proposed in this paper for automatic speech recognition (ASR).

26, TITLE:      Towards Robustifying NLI Models Against Lexical Dataset Biases
http://arxiv.org/abs/2005.04732
AUTHORS:        Xiang Zhou ; Mohit Bansal
COMMENTS:       ACL 2020 (13 pages)
HIGHLIGHT:      The first approach aims to remove the label bias at the embedding level.

27, TITLE:      Geometry-Aware Generation of Adversarial Point Clouds
http://arxiv.org/abs/1912.11171
AUTHORS:        Yuxin Wen ; Jiehong Lin ; Ke Chen ; C. L. Philip Chen ; Kui Jia
HIGHLIGHT:      Motivated by the different mechanisms when humans perceive 2D images and 3D shapes, we propose in this paper a new design of geometry-aware objectives, whose solutions favor (discrete versions of) the desired surface properties of smoothness and fairness.

28, TITLE:      Decidability of cutpoint isolation for probabilistic finite automata on letter-bounded inputs
http://arxiv.org/abs/2002.07660
AUTHORS:        Paul C. Bell ; Pavel Semukhin
COMMENTS:       17 pages
HIGHLIGHT:      We provide a constructive nondeterministic algorithm to solve the cutpoint isolation problem, which holds even when the PFA is exponentially ambiguous.

29, TITLE:      Sub-policy Adaptation for Hierarchical Reinforcement Learning
http://arxiv.org/abs/1906.05862
AUTHORS:        Alexander C. Li ; Carlos Florensa ; Ignasi Clavera ; Pieter Abbeel
COMMENTS:       ICLR 2020
HIGHLIGHT:      In this work, we propose a novel algorithm to discover a set of skills, and continuously adapt them along with the higher level even when training on a new task.

30, TITLE:      Artemis: A Novel Annotation Methodology for Indicative Single Document Summarization
http://arxiv.org/abs/2005.02146
AUTHORS:        Rahul Jha ; Keping Bi ; Yang Li ; Mahdi Pakdaman ; Asli Celikyilmaz ; Ivan Zhiboedov ; Kieran McDonald
HIGHLIGHT:      We describe Artemis (Annotation methodology for Rich, Tractable, Extractive, Multi-domain, Indicative Summarization), a novel hierarchical annotation process that produces indicative summaries for documents from multiple domains.

31, TITLE:      Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Fly
http://arxiv.org/abs/1911.03343
AUTHORS:        Nora Kassner ; Hinrich Schütze
HIGHLIGHT:      Building on Petroni et al. (2019), we propose two new probing tasks analyzing factual knowledge stored in Pretrained Language Models (PLMs).

32, TITLE:      Noise2Blur: Online Noise Extraction and Denoising
http://arxiv.org/abs/1912.01158
AUTHORS:        Huangxing Lin ; Weihong Zeng ; Xinghao Ding ; Xueyang Fu ; Yue Huang ; John Paisley
HIGHLIGHT:      We propose a new framework called Noise2Blur (N2B) for training robust image denoising models without pre-collected paired noisy/clean images.

33, TITLE:      Masked Language Model Scoring
http://arxiv.org/abs/1910.14659
AUTHORS:        Julian Salazar ; Davis Liang ; Toan Q. Nguyen ; Katrin Kirchhoff
COMMENTS:       Accepted to ACL 2020 (long paper)
HIGHLIGHT:      In all, PLLs and their associated pseudo-perplexities (PPPLs) enable plug-and-play use of the growing number of pretrained MLMs; e.g., we use a single cross-lingual model to rerank translations in multiple languages.

34, TITLE:      Verification of a Generative Separation Kernel
http://arxiv.org/abs/2001.10328
AUTHORS:        Inzemamul Haque ; Deepak D'Souza ; Habeeb P ; Arnab Kundu ; Ganesh Babu
HIGHLIGHT:      We propose a verification framework called conditional parametric refinement which allows us to formally reason about generative systems.

35, TITLE:      Microscopy Image Restoration with Deep Wiener-Kolmogorov filters
http://arxiv.org/abs/1911.10989
AUTHORS:        Valeriya Pronina ; Filippos Kokkinos ; Dmitry V. Dylov ; Stamatios Lefkimmiatis
COMMENTS:       Updated version
HIGHLIGHT:      In this work, we propose a unifying framework of algorithms for Gaussian image deblurring and denoising.

36, TITLE:      Human peripheral blur is optimal for object recognition
http://arxiv.org/abs/1807.08476
AUTHORS:        R. T. Pramod ; Harish Katti ; S. P. Arun
COMMENTS:       24 pages, 6 figures, 1 table
HIGHLIGHT:      To test this hypothesis, we trained deep neural networks on 'foveated' images with high resolution near objects and increasingly sparse sampling into the periphery.

37, TITLE:      WhiteNet: Zero-Day Phishing Website Detection by Visual Whitelists
http://arxiv.org/abs/1909.00300
AUTHORS:        Sahar Abdelnabi ; Katharina Krombholz ; Mario Fritz
HIGHLIGHT:      This paper contributes WhiteNet, a new similarity-based phishing detection framework, based on a triplet network with three shared Convolutional Neural Networks (CNNs).

38, TITLE:      Checkmate: Breaking the Memory Wall with Optimal Tensor Rematerialization
http://arxiv.org/abs/1910.02653
AUTHORS:        Paras Jain ; Ajay Jain ; Aniruddha Nrusimha ; Amir Gholami ; Pieter Abbeel ; Kurt Keutzer ; Ion Stoica ; Joseph E. Gonzalez
COMMENTS:       In Proceedings of 3rd Conference Machine Learning and Systems 2020 (MLSys 2020)
HIGHLIGHT:      We introduce Checkmate, a system that solves for optimal rematerialization schedules in reasonable times (under an hour) using off-the-shelf MILP solvers or near-optimal schedules with an approximation algorithm, then uses these schedules to accelerate millions of training iterations.

39, TITLE:      A High-Performance Object Proposals based on Horizontal High Frequency Signal
http://arxiv.org/abs/2003.06124
AUTHORS:        Jiang Chao ; Liang Huawei ; Wang Zhiling
HIGHLIGHT:      For this problem, we propose a class-independent object proposal algorithm BIHL.

40, TITLE:      Pose Proposal Critic: Robust Pose Refinement by Learning Reprojection Errors
http://arxiv.org/abs/2005.06262
AUTHORS:        Lucas Brynte ; Fredrik Kahl
COMMENTS:       Added acknowledgements
HIGHLIGHT:      In this paper we focus our attention on pose refinement, and show how to push the state-of-the-art further in the case of partial occlusions.

41, TITLE:      Transformer Networks for Trajectory Forecasting
http://arxiv.org/abs/2003.08111
AUTHORS:        Francesco Giuliari ; Irtiza Hasan ; Marco Cristani ; Fabio Galasso
COMMENTS:       18 pages, 3 figures
HIGHLIGHT:      We question the use of the LSTM models and propose the novel use of Transformer Networks for trajectory forecasting.

42, TITLE:      ESPRIT: Explaining Solutions to Physical Reasoning Tasks
http://arxiv.org/abs/2005.00730
AUTHORS:        Nazneen Fatema Rajani ; Rui Zhang ; Yi Chern Tan ; Stephan Zheng ; Jeremy Weiss ; Aadit Vyas ; Abhijit Gupta ; Caiming XIong ; Richard Socher ; Dragomir Radev
COMMENTS:       ACL 2020
HIGHLIGHT:      We propose ESPRIT, a framework for commonsense reasoning about qualitative physics in natural language that generates interpretable descriptions of physical events.

43, TITLE:      Non-dimensional Star-Identification
http://arxiv.org/abs/2003.13736
AUTHORS:        Carl Leake ; David Arnas ; Daniele Mortari
COMMENTS:       17 pages, 10 figures, 4 tables
HIGHLIGHT:      This study introduces a new "Non-Dimensional" star identification algorithm to reliably identify the stars observed by a wide field-of-view star tracker when the focal length and optical axis offset values are known with poor accuracy.

44, TITLE:      A Breezing Proof of the KMW Bound
http://arxiv.org/abs/2002.06005
AUTHORS:        Corinna Coupette ; Christoph Lenzen
COMMENTS:       21 pages, 6 figures
HIGHLIGHT:      Setting out to change this, in this work, we provide a fully self-contained and $\mathit{simple}$ proof of the KMW lower bound.

45, TITLE:      Open Loop In Natura Economic Planning
http://arxiv.org/abs/2005.01539
AUTHORS:        Spyridon Samothrakis
COMMENTS:       10 pages, 3 Figures
HIGHLIGHT:      Working within the tradition of Marx, Leontief, Kantorovich, Beer and Cockshott, we propose what we deem an automated planning system that aims to operate on unit level (e.g., factories and citizens), rather than on aggregate demand and sectors.

46, TITLE:      Fast in-place algorithms for polynomial operations: division, evaluation, interpolation
http://arxiv.org/abs/2002.10304
AUTHORS:        Pascal Giorgi ; Bruno Grenet ; Daniel S. Roche
HIGHLIGHT:      We demonstrate new in-place algorithms for the aforementioned polynomial computations which require only constant extra space and achieve the same asymptotic running time as their out-of-place counterparts.

47, TITLE:      A closed-form solution to estimate uncertainty in non-rigid structure from motion
http://arxiv.org/abs/2005.04810
AUTHORS:        Jingwei Song ; Mitesh Patel
COMMENTS:       8 pages, 2 figures
HIGHLIGHT:      In this paper, we present a statistical inference on the element-wise uncertainty quantification of the estimated deforming 3D shape points in the case of the exact low-rank SDP problem.

48, TITLE:      Anchors Based Method for Fingertips Position Estimation from a Monocular RGB Image using Deep Neural Network
http://arxiv.org/abs/2005.01351
AUTHORS:        Purnendu Mishra ; Kishor Sarawadekar
COMMENTS:       10 pages, 10 figures
HIGHLIGHT:      In this paper, we propose a deep neural network (DNN) based methodology to estimate the fingertips position.

49, TITLE:      Multi-label classification search space in the MEKA software
http://arxiv.org/abs/1811.11353
AUTHORS:        Alex G. C. de Sá ; Cristiano G. Pimenta ; Gisele L. Pappa ; Alex A. Freitas
COMMENTS:       Supplementary Material (GECCO'2020): Proposed Search Spaces
HIGHLIGHT:      Fundamentally, this occurs due to the problem transformation nature of several MLC algorithms used in this work.

50, TITLE:      Aquarium: Cassiopea and Alewife Languages
http://arxiv.org/abs/1908.00093
AUTHORS:        David A. Holland ; Jingmei Hu ; Ming Kawaguchi ; Eric Lu ; Stephen Chong ; Margo I. Seltzer
HIGHLIGHT:      This technical report describes two of the domain-specific languages used in the Aquarium kernel code synthesis project.

51, TITLE:      Generator From Edges: Reconstruction of Facial Images
http://arxiv.org/abs/2002.06682
AUTHORS:        Nao Takano ; Gita Alaghband
HIGHLIGHT:      We propose Generator From Edges (GFE) [Figure 2].

52, TITLE:      BLEURT: Learning Robust Metrics for Text Generation
http://arxiv.org/abs/2004.04696
AUTHORS:        Thibault Sellam ; Dipanjan Das ; Ankur P. Parikh
COMMENTS:       Accepted at ACL 2020
HIGHLIGHT:      We propose BLEURT, a learned evaluation metric based on BERT that can model human judgments with a few thousand possibly biased training examples.

53, TITLE:      A Locally Adaptive Interpretable Regression
http://arxiv.org/abs/2005.03350
AUTHORS:        Lkhagvadorj Munkhdalai ; Tsendsuren Munkhdalai ; Keun Ho Ryu
HIGHLIGHT:      In this work, we introduce a locally adaptive interpretable regression (LoAIR).

54, TITLE:      DeepFaceLab: A simple, flexible and extensible face swapping framework
http://arxiv.org/abs/2005.05535
AUTHORS:        Ivan Petrov ; Daiheng Gao ; Nikolay Chervoniy ; Kunlin Liu ; Sugasa Marangonda ; Chris Umé ; Mr. Dpfks ; Carl Shift Facenheim ; Luis RP ; Jian Jiang ; Sheng Zhang ; Pingyu Wu ; Bo Zhou ; Weiming Zhang
HIGHLIGHT:      In this paper, we detail the principles that drive the implementation of DeepFaceLab and introduce the pipeline of it, through which every aspect of the pipeline can be modified painlessly by users to achieve their customization purpose, and it's noteworthy that DeepFaceLab could achieve results with high fidelity and indeed indiscernible by mainstream forgery detection approaches.

55, TITLE:      An Early Study on Intelligent Analysis of Speech under COVID-19: Severity, Sleep Quality, Fatigue, and Anxiety
http://arxiv.org/abs/2005.00096
AUTHORS:        Jing Han ; Kun Qian ; Meishu Song ; Zijiang Yang ; Zhao Ren ; Shuo Liu ; Juan Liu ; Huaiyuan Zheng ; Wei Ji ; Tomoya Koike ; Xiao Li ; Zixing Zhang ; Yoshiharu Yamamoto ; Björn W. Schuller
HIGHLIGHT:      In this study, we focus on developing some potential use-cases of intelligent speech analysis for COVID-19 diagnosed patients.

56, TITLE:      Measuring Emotions in the COVID-19 Real World Worry Dataset
http://arxiv.org/abs/2004.04225
AUTHORS:        Bennett Kleinberg ; Isabelle van der Vegt ; Maximilian Mozes
COMMENTS:       Accepted to ACL 2020 COVID-19 workshop
HIGHLIGHT:      In this paper, we present the first ground truth dataset of emotional responses to COVID-19.

57, TITLE:      Extended 2D Consensus Hippocampus Segmentation
http://arxiv.org/abs/1902.04487
AUTHORS:        Diedre Carmo ; Bruna Silva ; Clarissa Yasuda ; Letícia Rittner ; Roberto Lotufo
COMMENTS:       This was published as an extended abstract in MIDL 2019 [arXiv:1907.08612]. An alpha version of the code is available at https://github.com/dscarmo/e2dhipseg. More experiments on improvements to the method and code are ongoing. Future updates are to be expected. A new, more complete paper is published in arXiv:2001.05058
HIGHLIGHT:      A method for volumetric hippocampus segmentation is presented, based on the consensus of tri-planar U-Net inspired fully convolutional networks (FCNNs), with some modifications, including residual connections, VGG weight transfers, batch normalization and a patch extraction technique employing data from neighbor patches.