2020.07.14.txt

==========New Papers==========
1, TITLE:       Probabilistic Jacobian-based Saliency Maps Attacks
http://arxiv.org/abs/2007.06032
AUTHORS:        António Loison ; Théo Combey ; Hatem Hajri
COMMENTS:       21 pages
HIGHLIGHT:      In this paper, we introduce Weighted JSMA (WJSMA) and Taylor JSMA (TJSMA), simple, faster and more efficient versions of JSMA.

2, TITLE:       Automatic Lyrics Transcription using Dilated Convolutional Neural Networks with Self-Attention
http://arxiv.org/abs/2007.06486
AUTHORS:        Emir Demirel ; Sven Ahlback ; Simon Dixon
HIGHLIGHT:      This paper proposes a complete pipeline for this task which may commonly be referred as automatic lyrics transcription (ALT).

3, TITLE:       Multi-Modality Information Fusion for Radiomics-based Neural Architecture Search
http://arxiv.org/abs/2007.06002
AUTHORS:        Yige Peng ; Lei Bi ; Michael Fulham ; Dagan Feng ; Jinman Kim
COMMENTS:       Accepted by MICCAI 2020
HIGHLIGHT:      In this study, we propose a multi-modality neural architecture search method (MM-NAS) to automatically derive optimal multi-modality image features for radiomics and thus negate the dependence on a manual process.

4, TITLE:       Usefulness of interpretability methods to explain deep learning based plant stress phenotyping
http://arxiv.org/abs/2007.05729
AUTHORS:        Koushik Nagasubramanian ; Asheesh K. Singh ; Arti Singh ; Soumik Sarkar ; Baskar Ganapathysubramanian
COMMENTS:       15 pages, 6 figures
HIGHLIGHT:      In this work, we compare some of the most popular interpretability methods: Saliency Maps, SmoothGrad, Guided Backpropogation, Deep Taylor Decomposition, Integrated Gradients, Layer-wise Relevance Propagation and Gradient times Input, for interpreting the deep learning model.

5, TITLE:       TERA: Self-Supervised Learning of Transformer Encoder Representation for Speech
http://arxiv.org/abs/2007.06028
AUTHORS:        Andy T. Liu ; Shang-Wen Li ; Hung-yi Lee
COMMENTS:       Submitted to IEEE TASLP, currently under review
HIGHLIGHT:      We introduce a self-supervised speech pre-training method called TERA, which stands for Transformer Encoder Representations from Alteration.

6, TITLE:       Accelerating Translational Image Registration for HDR Images on GPU
http://arxiv.org/abs/2007.06483
AUTHORS:        Kadir Cenk Alpay ; Kadir Berkay Aydemir ; Alptekin Temizel
COMMENTS:       Submitted for Consideration for Publication in High Performance Computing Conference 2020
HIGHLIGHT:      In this study, we optimize these computations using a parallel processing approach utilizing GPU.

7, TITLE:       MeDaS: An open-source platform as service to help break the walls between medicine and informatics
http://arxiv.org/abs/2007.06013
AUTHORS:        Liang Zhang ; Johann Li ; Ping Li ; Xiaoyuan Lu ; Peiyi Shen ; Guangming Zhu ; Syed Afaq Shah ; Mohammed Bennarmoun ; Kun Qian ; Björn W. Schuller
HIGHLIGHT:      Based on a series of toolkits and utilities from the idea of RINV (Rapid Implementation aNd Verification), our proposed MeDaS platform can implement pre-processing, post-processing, augmentation, visualization, and other phases needed in medical image analysis.

8, TITLE:       HSD Shared Task in VLSP Campaign 2019:Hate Speech Detection for Social Good
http://arxiv.org/abs/2007.06493
AUTHORS:        Xuan-Son Vu ; Thanh Vu ; Mai-Vu Tran ; Thanh Le-Cong ; Huyen T M. Nguyen
HIGHLIGHT:      The paper describes the organisation of the "HateSpeech Detection" (HSD) task at the VLSP workshop 2019 on detecting the fine-grained presence of hate speech in Vietnamese textual items (i.e., messages) extracted from Facebook, which is the most popular social network site (SNS) in Vietnam.

9, TITLE:       Distributed Graph Convolutional Networks
http://arxiv.org/abs/2007.06281
AUTHORS:        Simone Scardapane ; Indro Spinelli ; Paolo Di Lorenzo
COMMENTS:       Preprint submitted to IEEE TSIPN
HIGHLIGHT:      The aim of this work is to develop a fully-distributed algorithmic framework for training graph convolutional networks (GCNs).

10, TITLE:      Strengthening neighbourhood substitution
http://arxiv.org/abs/2007.06282
AUTHORS:        Martin C. Cooper
HIGHLIGHT:      We show that the notion of neighbourhood substitution can be strengthened in two distinct ways without increasing time complexity.

11, TITLE:      Term Revealing: Furthering Quantization at Run Time on Quantized DNNs
http://arxiv.org/abs/2007.06389
AUTHORS:        H. T. Kung ; Bradley McDanel ; Sai Qian Zhang
COMMENTS:       13 pages, 19 figures, 4 tables, To appear in Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC), 2020
HIGHLIGHT:      We present a novel technique, called Term Revealing (TR), for furthering quantization at run time for improved performance of Deep Neural Networks (DNNs) already quantized with conventional quantization methods.

12, TITLE:      A simple defense against adversarial attacks on heatmap explanations
http://arxiv.org/abs/2007.06381
AUTHORS:        Laura Rieger ; Lars Kai Hansen
COMMENTS:       Accepted at 2020 Workshop on Human Interpretability in Machine Learning (WHI)
HIGHLIGHT:      In our work we present an effective defence against such adversarial attacks on neural networks.

13, TITLE:      Knowledge Graph Driven Approach to Represent Video Streams for Spatiotemporal Event Pattern Matching in Complex Event Processing
http://arxiv.org/abs/2007.06292
AUTHORS:        Piyush Yadav ; Dhaval Salwala ; Edward Curry
COMMENTS:       31 pages, 14 Figures, Publication accepted in International Journal of Graph Computing
HIGHLIGHT:      This work introduces a graph-based structure for continuous evolving video streams, which enables the CEP system to query complex video event patterns.

14, TITLE:      Seeing Eye-to-Eye? A Comparison of Object Recognition Performance in Humans and Deep Convolutional Neural Networks under Image Manipulation
http://arxiv.org/abs/2007.06294
AUTHORS:        Leonard E. van Dyck ; Walter R. Gruber
COMMENTS:       19 pages, 7 figures, 3 tables
HIGHLIGHT:      This study aims towards a behavioral comparison of visual core object recognition between humans and feedforward neural networks in a classification learning paradigm on an ImageNet data set.

15, TITLE:      Paranoid Transformer: Reading Narrative of Madness as Computational Approach to Creativity
http://arxiv.org/abs/2007.06290
AUTHORS:        Yana Agafonova ; Alexey Tikhonov ; Ivan P. Yamshchikov
HIGHLIGHT:      It presents a case study of a Paranoid Transformer - a fully autonomous text generation engine with raw output that could be read as the narrative of a mad digital persona without any additional human post-filtering.

16, TITLE:      DeepHAZMAT: Hazardous Materials Sign Detection and Segmentation with Restricted Computational Resources
http://arxiv.org/abs/2007.06392
AUTHORS:        Amir Sharifi ; Ahmadreza Zibaei ; Mahdi Rezaei
HIGHLIGHT:      In this paper, we propose a CNN-Based pipeline called DeepHAZMAT for detecting and segmenting Hazmats in four steps; 1) optimising the number of input images that are fed into the CNN network, 2) using the YOLOv3-tiny structure to collect the required visual information from the hazardous areas, 3) Hazmat sign segmentation and separation from the background using GrabCut technique, and 4) post-processing the result with morphological operators and convex hall algorithm.

17, TITLE:      A Feature Analysis for Multimodal News Retrieval
http://arxiv.org/abs/2007.06390
AUTHORS:        Golsa Tahmasebzadeh ; Sherzod Hakimov ; Eric Müller-Budack ; Ralph Ewerth
HIGHLIGHT:      In this paper, we investigate the usefulness of multimodal features for cross-lingual news search in various domains: politics, health, environment, sport, and finance.

18, TITLE:      BoxE: A Box Embedding Model for Knowledge Base Completion
http://arxiv.org/abs/2007.06267
AUTHORS:        Ralph Abboud ; İsmail İlkan Ceylan ; Thomas Lukasiewicz ; Tommaso Salvatori
HIGHLIGHT:      Here, we propose a spatio-translational embedding model, called BoxE, that simultaneously addresses all these limitations.

19, TITLE:      Contextual Bandit with Missing Rewards
http://arxiv.org/abs/2007.06368
AUTHORS:        Djallel Bouneffouf ; Sohini Upadhyay ; Yasaman Khazaeni
HIGHLIGHT:      In order to address the missing rewards setting, we propose to combine the standard contextual bandit approach with an unsupervised learning mechanism such as clustering.

20, TITLE:      OpenStreetMap: Challenges and Opportunities in Machine Learning and Remote Sensing
http://arxiv.org/abs/2007.06277
AUTHORS:        John Vargas ; Shivangi Srivastava ; Devis Tuia ; Alexandre Falcao
HIGHLIGHT:      In this work, we present a review of recent methods based on machine learning to improve and use OSM data.

21, TITLE:      On uncertainty estimation in active learning for image segmentation
http://arxiv.org/abs/2007.06364
AUTHORS:        Bo Li ; Tommy Sonne Alstrøm
COMMENTS:       Presented at ICML 2020 Workshop on Uncertainty & Robustness in Deep Learning
HIGHLIGHT:      In this paper, we explore uncertainty calibration within an active learning framework for medical image segmentation, an area where labels often are scarce.

22, TITLE:      Dual-Teacher: Integrating Intra-domain and Inter-domain Teachers for Annotation-efficient Cardiac Segmentation
http://arxiv.org/abs/2007.06279
AUTHORS:        Kang Li ; Shujun Wang ; Lequan Yu ; Pheng-Ann Heng
COMMENTS:       Accepted at MICCAI 2020
HIGHLIGHT:      In this paper, we aim to investigate the feasibility of simultaneously leveraging abundant unlabeled data and well-established cross-modality data for annotation-efficient medical image segmentation.

23, TITLE:      RNA-2QCFA: Evolving Two-way Quantum Finite Automata with Classical States for RNA Secondary Structures
http://arxiv.org/abs/2007.06273
AUTHORS:        Amandeep Singh Bhatia ; Shenggen Zheng
COMMENTS:       9 pages, 3 figures
HIGHLIGHT:      The main objective of this paper is on using two-way quantum finite automata with classical states to simulate, model and analyze the ribonucleic acid (RNA) sequences.

24, TITLE:      RATT: Recurrent Attention to Transient Tasks for Continual Image Captioning
http://arxiv.org/abs/2007.06271
AUTHORS:        Riccardo Del Chiaro ; Bartłomiej Twardowski ; Andrew D. Bagdanov ; Joost van de Weijer
COMMENTS:       8 pages, 5 figures, to be published in LifelongML workshop at ICML2020
HIGHLIGHT:      In this paper we take a systematic look at continual learning of LSTM-based models for image captioning.

25, TITLE:      Screen Tracking for Clinical Translation of Live Ultrasound Image Analysis Methods
http://arxiv.org/abs/2007.06272
AUTHORS:        Simona Treivase ; Alberto Gomez ; Jacqueline Matthew ; Emily Skelton ; Julia A. Schnabel ; Nicolas Toussaint
HIGHLIGHT:      We propose a generic framework to extract the US images and superimpose the results of an analysis task, without any need for physical connection or alteration to the US system.

26, TITLE:      Smart technology in the classroom: a systematic review.Prospects for algorithmic accountability
http://arxiv.org/abs/2007.06374
AUTHORS:        Arian Garshi ; Malin Wist Jakobsen ; Jørgen Nyborg-Christensen ; Daniel Ostnes ; Maria Ovchinnikova
HIGHLIGHT:      Based on our insights we propose a framework to effectively identify accountability for smart technology in education.

27, TITLE:      Symmetric Dilated Convolution for Surgical Gesture Recognition
http://arxiv.org/abs/2007.06373
AUTHORS:        Jinglu Zhang ; Yinyu Nie ; Yao Lyu ; Hailin Li ; Jian Chang ; Xiaosong Yang ; Jian Jun Zhang
COMMENTS:       Accepted to MICCAI 2020
HIGHLIGHT:      To tackle these challenges, we propose a novel temporal convolutional architecture to automatically detect and segment surgical gestures with corresponding boundaries only using RGB videos.

28, TITLE:      Expert Training: Task Hardness Aware Meta-Learning for Few-Shot Classification
http://arxiv.org/abs/2007.06240
AUTHORS:        Yucan Zhou ; Yu Wang ; Jianfei Cai ; Yu Zhou ; Qinghua Hu ; Weiping Wang
COMMENTS:       9 pages, 6 figures
HIGHLIGHT:      Inspired by this idea, we propose an easy-to-hard expert meta-training strategy to arrange the training tasks properly, where easy tasks are preferred in the first phase, then, hard tasks are emphasized in the second phase.

29, TITLE:      Transformer with Depth-Wise LSTM
http://arxiv.org/abs/2007.06257
AUTHORS:        Hongfei Xu ; Qiuhui Liu ; Deyi Xiong ; Josef van Genabith
HIGHLIGHT:      In this paper, we suggest that the residual connection has its drawbacks, and propose to train Transformers with the depth-wise LSTM which regards outputs of layers as steps in time series instead of residual connections, under the motivation that the vanishing gradient problem suffered by deep networks is the same as recurrent networks applied to long sequences, while LSTM (Hochreiter and Schmidhuber, 1997) has been proven of good capability in capturing long-distance relationship, and its design may alleviate some drawbacks of residual connections while ensuring the convergence.

30, TITLE:      A theory of interaction semantics
http://arxiv.org/abs/2007.06258
AUTHORS:        Johannes Reich
HIGHLIGHT:      The aim of this article is to delineate a theory of interaction semantics and thereby provide a proper understanding of the "meaning" of the exchanged characters within an interaction.

31, TITLE:      Exploring the Evolution of GANs through Quality Diversity
http://arxiv.org/abs/2007.06251
AUTHORS:        Victor Costa ; Nuno Lourenço ; João Correia ; Penousal Machado
COMMENTS:       Published in GECCO 2020
HIGHLIGHT:      We propose in this paper the application of a quality-diversity algorithm in the evolution of GANs.

32, TITLE:      Locality Guided Neural Networks for Explainable Artificial Intelligence
http://arxiv.org/abs/2007.06131
AUTHORS:        Randy Tan ; Naimul Khan ; Ling Guan
COMMENTS:       8 pages, 3 figures, submitted to WCCI2020
HIGHLIGHT:      In this paper, we propose a novel algorithm for back propagation, called Locality Guided Neural Network(LGNN) for training networks that preserves locality between neighbouring neurons within each layer of a deep network.

33, TITLE:      DRWR: A Differentiable Renderer without Rendering for Unsupervised 3D Structure Learning from Silhouette Images
http://arxiv.org/abs/2007.06127
AUTHORS:        Zhizhong Han ; Chao Chen ; Yu-Shen Liu ; Matthias Zwicker
COMMENTS:       Accepted at ICML2020
HIGHLIGHT:      In contrast, here we propose a Differentiable Renderer Without Rendering (DRWR) that omits these steps.

34, TITLE:      SkyScapes -- Fine-Grained Semantic Understanding of Aerial Scenes
http://arxiv.org/abs/2007.06102
AUTHORS:        Seyed Majid Azimi ; Corentin Henry ; Lars Sommer ; Arne Schumann ; Eleonora Vig
COMMENTS:       Accepted in IEEE ICCV19
HIGHLIGHT:      We therefore propose a novel multi-task model, which incorporates semantic edge detection and is better tuned for feature extraction from a wide range of scales.

35, TITLE:      VINNAS: Variational Inference-based Neural Network Architecture Search
http://arxiv.org/abs/2007.06103
AUTHORS:        Martin Ferianc ; Hongxiang Fan ; Miguel Rodrigues
COMMENTS:       Submitted to ICPR'20 https://github.com/iiml-ucl/vinnas
HIGHLIGHT:      To address these defects, we present a differentiable variational inference-based NAS method for searching sparse convolutional neural networks.

36, TITLE:      Neural disambiguation of lemma and part of speech in morphologically rich languages
http://arxiv.org/abs/2007.06104
AUTHORS:        José María Hoya Quecedo ; Maximilian W. Koppatz ; Giacomo Furlan ; Roman Yangarber
COMMENTS:       This paper contains corrigenda to a previously published paper (Hoya Quecedo et al., 2020). It corrects a mistake in the original evaluation setup, and the results reported in Section 6., in Tables 5, 6, and 7
HIGHLIGHT:      We propose a method for disambiguating ambiguous words in context, using a large un-annotated corpus of text, and a morphological analyser -- with no manual disambiguation or data annotation.

37, TITLE:      Tabletop Roleplaying Games as Procedural Content Generators
http://arxiv.org/abs/2007.06108
AUTHORS:        Matthew Guzdial ; Devi Acharya ; Max Kreminski ; Michael Cook ; Mirjam Eladhari ; Antonios Liapis ; Anne Sullivan
COMMENTS:       9 pages, 2 figures, FDG Workshop on Procedural Content Generation 2020
HIGHLIGHT:      In this paper, we argue that TTRPG design can usefully be viewed as procedural content generator design.

38, TITLE:      Graph Structure of Neural Networks
http://arxiv.org/abs/2007.06559
AUTHORS:        Jiaxuan You ; Jure Leskovec ; Kaiming He ; Saining Xie
COMMENTS:       ICML 2020
HIGHLIGHT:      Here we systematically investigate how does the graph structure of neural networks affect their predictive performance.

39, TITLE:      Long-Term Planning with Deep Reinforcement Learning on Autonomous Drones
http://arxiv.org/abs/2007.05694
AUTHORS:        Ugurkan Ates
COMMENTS:       Submitted to Association for the Advancement of Artificial Intelligence(AAAI) 2020 Fall Symposium Series
HIGHLIGHT:      In this paper, we study a long-term planning scenario that is based on drone racing competitions held in real life.

40, TITLE:      Free-running SIMilarity-Based Angiography (SIMBA) for simplified anatomical MR imaging of the heart
http://arxiv.org/abs/2007.06544
AUTHORS:        John Heerfordt ; Kevin K. Whitehead ; Jessica A. M. Bastiaansen ; Lorenzo Di Sopra ; Christopher W. Roy ; Jérôme Yerly ; Bastien Milani ; Mark A. Fogel ; Matthias Stuber ; Davide Piccini
COMMENTS:       8 figures, 2 tables
HIGHLIGHT:      We propose a novel fast reconstruction algorithm, applicable to ungated free-running sequences, that leverages inherent similarities in the acquired data to avoid such physiological constraints.

41, TITLE:      Fast Video Object Segmentation With Temporal Aggregation Network and Dynamic Template Matching
http://arxiv.org/abs/2007.05687
AUTHORS:        Xuhua Huang ; Jiarui Xu ; Yu-Wing Tai ; Chi-Keung Tang
COMMENTS:       CVPR2020
HIGHLIGHT:      In this paper, we introduce "tracking-by-detection" into VOS which can coherently integrate segmentation into tracking, by proposing a new temporal aggregation network and a novel dynamic time-evolving template matching mechanism to achieve significantly improved performance.

42, TITLE:      Semi-steady-state Jaya Algorithm
http://arxiv.org/abs/2007.06463
AUTHORS:        Uday K. Chakraborty
HIGHLIGHT:      The present paper proposes a new, improved Jaya algorithm by modifying the update strategies of the best and the worst members in the population.

43, TITLE:      Automatic Pass Annotation from Soccer VideoStreams Based on Object Detection and LSTM
http://arxiv.org/abs/2007.06475
AUTHORS:        Danilo Sorano ; Fabio Carrara ; Paolo Cintia ; Fabrizio Falchi ; Luca Pappalardo
HIGHLIGHT:      In this paper, we describe PassNet, a method to recognize the most frequent events in soccer, i.e., passes, from video streams.

44, TITLE:      Learning Reasoning Strategies in End-to-End Differentiable Proving
http://arxiv.org/abs/2007.06477
AUTHORS:        Pasquale Minervini ; Sebastian Riedel ; Pontus Stenetorp ; Edward Grefenstette ; Tim Rocktäschel
COMMENTS:       Proceedings of the 37th International Conference on Machine Learning (ICML 2020)
HIGHLIGHT:      We present Conditional Theorem Provers (CTPs), an extension to NTPs that learns an optimal rule selection strategy via gradient-based optimisation.

45, TITLE:      Implicit Euler ODE Networks for Single-Image Dehazing
http://arxiv.org/abs/2007.06443
AUTHORS:        Jiawei Shen ; Zhuoyan Li ; Lei Yu ; Gui-Song Xia ; Wen Yang
COMMENTS:       10pages, 10 figures, "for the associate project, see https://github.com/Jiawei-Shen?tab=repositories", submitted to CVPR workshop "vision for four seasons",
HIGHLIGHT:      In this paper, we extend the explicit forward approximation to the implicit backward counterpart, which can be realized via a recursive neural network, named IM-block.

46, TITLE:      Attention-guided Quality Assessment for Automated Cryo-EM Grid Screening
http://arxiv.org/abs/2007.05593
AUTHORS:        Hong Xu ; Shireen Y. Elhabian ; David E. Timm
COMMENTS:       Accepted for publication in MICCAI 2020, the 23rd International Conference on Medical Image Computing and Computer Assisted Intervention
HIGHLIGHT:      Here, we focus on automating the early decision making for the microscope operator, scoring low magnification images of squares, and proposing the first deep learning framework, XCryoNet, for automated cryo-EM grid screening.

47, TITLE:      EMIXER: End-to-end Multimodal X-ray Generation via Self-supervision
http://arxiv.org/abs/2007.05597
AUTHORS:        Siddharth Biswal ; Peiye Zhuang ; Ayis Pyrros ; Nasir Siddiqui ; Sanmi Koyejo ; Jimeng Sun
HIGHLIGHT:      To tackle this joint synthesis challenge, we propose an End-to-end MultImodal X-ray genERative model (EMIXER) for jointly synthesizing x-ray images and corresponding free-text reports, all conditional on diagnosis labels.

48, TITLE:      Whitening for Self-Supervised Representation Learning
http://arxiv.org/abs/2007.06346
AUTHORS:        Aleksandr Ermolov ; Aliaksandr Siarohin ; Enver Sangineto ; Nicu Sebe
HIGHLIGHT:      In this paper we propose a different direction and a new loss function for self-supervised learning which is based on the whitening of the latent-space features.

49, TITLE:      End-to-End Multi-Object Tracking with Global Response Map
http://arxiv.org/abs/2007.06344
AUTHORS:        Xingyu Wan ; Jiakai Cao ; Sanping Zhou ; Jinjun Wang
HIGHLIGHT:      To address the problem, we present a completely end-to-end approach that takes image-sequence/video as input and outputs directly the located and tracked objects of learned types.

50, TITLE:      Learning and Exploiting Interclass Visual Correlations for Medical Image Classification
http://arxiv.org/abs/2007.06371
AUTHORS:        Dong Wei ; Shilei Cao ; Kai Ma ; Yefeng Zheng
HIGHLIGHT:      In this paper, we present the Class-Correlation Learning Network (CCL-Net) to learn interclass visual correlations from given training data, and produce soft labels to help with classification tasks.

51, TITLE:      DeU-Net: Deformable U-Net for 3D Cardiac MRI Video Segmentation
http://arxiv.org/abs/2007.06341
AUTHORS:        Shunjie Dong ; Jinlong Zhao ; Maojun Zhang ; Zhengxue Shi ; Jianing Deng ; Yiyu Shi ; Mei Tian ; Cheng Zhuo
HIGHLIGHT:      In this paper, we propose a novel Deformable U-Net (DeU-Net) to fully exploit spatio-temporal information from 3D cardiac MRI video, including a Temporal Deformable Aggregation Module (TDAM) and a Deformable Global Position Attention (DGPA) network.

52, TITLE:      A Label Attention Model for ICD Coding from Clinical Text
http://arxiv.org/abs/2007.06351
AUTHORS:        Thanh Vu ; Dat Quoc Nguyen ; Anthony Nguyen
COMMENTS:       In Proceedings of IJCAI 2020 (Main Track)
HIGHLIGHT:      In this paper, we propose a new label attention model for automatic ICD coding, which can handle both the various lengths and the interdependence of the ICD code related text fragments.

53, TITLE:      Disentanglement of Color and Shape Representations for Continual Learning
http://arxiv.org/abs/2007.06356
AUTHORS:        David Berga ; Marc Masana ; Joost Van de Weijer
COMMENTS:       Accepted at CL-ICML 2020
HIGHLIGHT:      We combine our method with Elastic Weight Consolidation, Learning without Forgetting, Synaptic Intelligence and Memory Aware Synapses, and show that feature disentanglement positively impacts continual learning performance.

54, TITLE:      Multiple Sound Sources Localization from Coarse to Fine
http://arxiv.org/abs/2007.06355
AUTHORS:        Rui Qian ; Di Hu ; Heinrich Dinkel ; Mengyue Wu ; Ning Xu ; Weiyao Lin
COMMENTS:       to appear in ECCV 2020
HIGHLIGHT:      To solve this problem, we develop a two-stage audiovisual learning framework that disentangles audio and visual representations of different categories from complex scenes, then performs cross-modal feature alignment in a coarse-to-fine manner.

55, TITLE:      ProtTrans: Towards Cracking the Language of Life's Code Through Self-Supervised Deep Learning and High Performance Computing
http://arxiv.org/abs/2007.06225
AUTHORS:        Ahmed Elnaggar ; Michael Heinzinger ; Christian Dallago ; Ghalia Rihawi ; Yu Wang ; Llion Jones ; Tom Gibbs ; Tamas Feher ; Christoph Angerer ; Debsindhu Bhowmik ; Burkhard Rost
HIGHLIGHT:      Methodology: Here, we trained two auto-regressive language models (Transformer-XL and XLNet) and two auto-encoder models (BERT and Albert) using 80 billion amino acids from 200 million protein sequences (UniRef100) and 393 billion amino acids from 2.1 billion protein sequences (BFD).

56, TITLE:      Location-Aware Box Reasoning for Anchor-Based Single-Shot Object Detection
http://arxiv.org/abs/2007.06233
AUTHORS:        Wenchi Ma ; Kaidong Li ; Guanghui Wang
HIGHLIGHT:      In this paper, we aim at single-shot object detectors and propose a location-aware anchor-based reasoning (LAAR) for the bounding boxes.

57, TITLE:      Active Crowd Counting with Limited Supervision
http://arxiv.org/abs/2007.06334
AUTHORS:        Zhen Zhao ; Miaojing Shi ; Xiaoxiao Zhao ; Li Li
COMMENTS:       ECCV2020 camera ready
HIGHLIGHT:      Active Crowd Counting with Limited Supervision

58, TITLE:      IntegralAction: Pose-driven Feature Integration for Robust Human Action Recognition in Videos
http://arxiv.org/abs/2007.06317
AUTHORS:        Gyeongsik Moon ; Heeseung Kwon ; Kyoung Mu Lee ; Minsu Cho
HIGHLIGHT:      To address this problem, we propose to learn pose-driven feature integration that dynamically combines appearance and pose streams by observing pose features on the fly.

59, TITLE:      Hierarchical Dynamic Filtering Network for RGB-D Salient Object Detection
http://arxiv.org/abs/2007.06227
AUTHORS:        Youwei Pang ; Lihe Zhang ; Xiaoqi Zhao ; Huchuan Lu
COMMENTS:       The work has been accepted by ECCV 2020
HIGHLIGHT:      In this paper, we explore these issues from a new perspective.

60, TITLE:      Structured Policy Iteration for Linear Quadratic Regulator
http://arxiv.org/abs/2007.06202
AUTHORS:        Youngsuk Park ; Ryan A. Rossi ; Zheng Wen ; Gang Wu ; Handong Zhao
HIGHLIGHT:      In this paper, we introduce the \textit{Structured Policy Iteration} (S-PI) for LQR, a method capable of deriving a structured linear policy.

61, TITLE:      Generating Functions for Probabilistic Programs
http://arxiv.org/abs/2007.06327
AUTHORS:        Lutz Klinkenberg ; Kevin Batz ; Benjamin Lucien Kaminski ; Joost-Pieter Katoen ; Joshua Moerman ; Tobias Winkler
HIGHLIGHT:      This paper investigates the usage of generating functions (GFs) encoding measures over the program variables for reasoning about discrete probabilistic programs.

62, TITLE:      Domain aware medical image classifier interpretation by counterfactual impact analysis
http://arxiv.org/abs/2007.06312
AUTHORS:        Dimitrios Lenis ; David Major ; Maria Wimmer ; Astrid Berg ; Gert Sluiter ; Katja Bühler
COMMENTS:       Accepted for publication at International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2020
HIGHLIGHT:      In this work, we discuss and overcome these obstacles by introducing a neural-network based attribution method, applicable to any trained predictor.

63, TITLE:      Part-aware Prototype Network for Few-shot Semantic Segmentation
http://arxiv.org/abs/2007.06309
AUTHORS:        Yongfei Liu ; Xiangyi Zhang ; Songyang Zhang ; Xuming He
COMMENTS:       ECCV-2020
HIGHLIGHT:      In this paper, we propose a novel few-shot semantic segmentation framework based on the prototype representation.

64, TITLE:      DinerDash Gym: A Benchmark for Policy Learning in High-Dimensional Action Space
http://arxiv.org/abs/2007.06207
AUTHORS:        Siwei Chen ; Xiao Ma ; David Hsu
HIGHLIGHT:      On top of that, we introduce Decomposed Policy Graph Modelling (DPGM), an algorithm that combines both graph modelling and deep learning to allow explicit domain knowledge embedding and achieves significant improvement comparing to the baseline.

65, TITLE:      Decoupling Inherent Risk and Early Cancer Signs in Image-based Breast Cancer Risk Models
http://arxiv.org/abs/2007.05791
AUTHORS:        Yue Liu ; Hossein Azizpour ; Fredrik Strand ; Kevin Smith
COMMENTS:       Medical Image Computing and Computer Assisted Interventions 2020
HIGHLIGHT:      With this in mind, we trained networks using three different criteria to select the positive training data (i.e. images from patients that will develop cancer): an inherent risk model trained on images with no visible signs of cancer, a cancer signs model trained on images containing cancer or early signs of cancer, and a conflated model trained on all images from patients with a cancer diagnosis.

66, TITLE:      Generalization of Deep Convolutional Neural Networks -- A Case-study on Open-source Chest Radiographs
http://arxiv.org/abs/2007.05786
AUTHORS:        Nazanin Mashhaditafreshi ; Amara Tariq ; Judy Wawira Gichoya ; Imon Banerjee
HIGHLIGHT:      We use InceptionResNetV2 and DenseNet121 architectures to predict the risk of 5 common chest pathologies.

67, TITLE:      Planning on the fast lane: Learning to interact using attention mechanisms in path integral inverse reinforcement learning
http://arxiv.org/abs/2007.05798
AUTHORS:        Sascha Rosbach ; Xing Li ; Simon Großjohann ; Silviu Homoceanu ; Stefan Roth
COMMENTS:       Manuscript accepted at 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Received March 1, 2020. Typos corrected
HIGHLIGHT:      In this work, we are concerned with the sequential reward prediction over an extended time horizon.

68, TITLE:      Illuminating Mario Scenes in the Latent Space of a Generative Adversarial Network
http://arxiv.org/abs/2007.05674
AUTHORS:        Matthew C. Fontaine ; Ruilin Liu ; Julian Togelius ; Amy K. Hoover ; Stefanos Nikolaidis
HIGHLIGHT:      We introduce a new method called latent space illumination (LSI), which uses state-of-the-art quality diversity algorithms designed to optimize in continuous spaces, i.e., MAP-Elites with a directional variation operator and Covariance Matrix Adaptation MAP-Elites, to effectively search the parameter space of theGAN along a set of multiple level mechanics.

69, TITLE:      Coarse-to-Fine Pseudo-Labeling Guided Meta-Learning for Few-Shot Classification
http://arxiv.org/abs/2007.05675
AUTHORS:        Jinhai Yang ; Hua Yang ; Lin Chen
HIGHLIGHT:      In this paper, we show that meta-learning models can extract transferable knowledge from coarse-grained supervision for few-shot classification.

70, TITLE:      Uncertain-DeepSSM: From Images to Probabilistic Shape Models
http://arxiv.org/abs/2007.06516
AUTHORS:        Jadie Adams ; Riddhish Bhalodia ; Shireen Elhabian
COMMENTS:       16 pages, 7 figures
HIGHLIGHT:      Here, we propose Uncertain-DeepSSM as a unified model that quantifies both, data-dependent aleatoric uncertainty by adapting the network to predict intrinsic input variance, and model-dependent epistemic uncertainty via a Monte Carlo dropout sampling to approximate a variational distribution over the network parameters.

71, TITLE:      To filter prune, or to layer prune, that is the question
http://arxiv.org/abs/2007.05667
AUTHORS:        Sara Elkerdawy ; Mostafa Elhoushi ; Abhineet Singh ; Hong Zhang ; Nilanjan Ray
HIGHLIGHT:      In this paper, we show the limitation of filter pruning methods in terms of latency reduction and propose LayerPrune framework.

72, TITLE:      Fast Real-time Counterfactual Explanations
http://arxiv.org/abs/2007.05684
AUTHORS:        Yunxia Zhao
COMMENTS:       This paper has been accepted by ICML workshop 2020
HIGHLIGHT:      Counterfactual explanations are considered, which is to answer {\it why the prediction is class A but not B.} Different from previous optimization based methods, an optimization-free Fast ReAl-time Counterfactual Explanation (FRACE) algorithm is proposed benefiting from the development of multi-domain image to image translation algorithms.

73, TITLE:      Feature Selection on Noisy Twitter Short Text Messages for Language Identification
http://arxiv.org/abs/2007.05727
AUTHORS:        Mohd Zeeshan Ansari ; Tanvir Ahmad ; Ana Fatima
HIGHLIGHT:      In this article, we basically consider the Hindi-English language identification task as Hindi and English are often two most widely spoken languages of India.

74, TITLE:      ECML: An Ensemble Cascade Metric Learning Mechanism towards Face Verification
http://arxiv.org/abs/2007.05720
AUTHORS:        Fu Xiong ; Yang Xiao ; Zhiguo Cao ; Yancheng Wang ; Joey Tianyi Zhou ; Jianxi Wu
COMMENTS:       Accepted to IEEE Transaction on Cybernetics
HIGHLIGHT:      Hence, we propose a novel ensemble cascade metric learning (ECML) mechanism.

75, TITLE:      Do We Need Sound for Sound Source Localization?
http://arxiv.org/abs/2007.05722
AUTHORS:        Takashi Oya ; Shohei Iwase ; Ryota Natsume ; Takahiro Itazuri ; Shugo Yamaguchi ; Shigeo Morishima
COMMENTS:       Paper: 14 pages, 6 figures. Supplementary Material: 6 pages, 3 figures. Videos and Codes will be released later
HIGHLIGHT:      As an alternative, we present an evaluation protocol that enforces both visual and aural information to be leveraged, and verify this property through several experiments.

76, TITLE:      Enhanced Behavioral Cloning Based self-driving Car Using Transfer Learning
http://arxiv.org/abs/2007.05740
AUTHORS:        Uppala Sumanth ; Narinder Singh Punn ; Sanjay Kumar Sonbhadra ; Sonali Agarwal
HIGHLIGHT:      Concerning the same, the present paper proposes a transfer learning approach using VGG16 architecture, which is fine tuned by retraining the last block while keeping other blocks as non-trainable.

77, TITLE:      Learning Object Depth from Camera Motion and Video Object Segmentation
http://arxiv.org/abs/2007.05676
AUTHORS:        Brent A. Griffin ; Jason J. Corso
HIGHLIGHT:      We demonstrate our approach across domains using a robot camera to locate objects from the YCB dataset and a vehicle camera to locate obstacles while driving.

78, TITLE:      Deep or Simple Models for Semantic Tagging? It Depends on your Data [Experiments]
http://arxiv.org/abs/2007.05651
AUTHORS:        Jinfeng Li ; Yuliang Li ; Xiaolan Wang ; Wang-Chiew Tan
HIGHLIGHT:      To answer this question, we compare deep models against "simple models" over datasets with varying characteristics.

79, TITLE:      Learning Local Complex Features using Randomized Neural Networks for Texture Analysis
http://arxiv.org/abs/2007.05643
AUTHORS:        Lucas C. Ribas ; Leonardo F. S. Scabini ; Jarbas Joaci de Mesquita Sá Junior ; Odemir M. Bruno
HIGHLIGHT:      In this paper, we present a new approach that combines a learning technique and the Complex Network (CN) theory for texture analysis.

80, TITLE:      Cascade Network with Guided Loss and Hybrid Attention for Two-view Geometry
http://arxiv.org/abs/2007.05706
AUTHORS:        Zhi Chen ; Fan Yang Wenbing Tao
HIGHLIGHT:      In this paper, we are committed to designing a high-performance network for two-view geometry.

81, TITLE:      Deep Patch-based Human Segmentation
http://arxiv.org/abs/2007.05661
AUTHORS:        Dongbo Zhang ; Zheng Fang ; Xuequan Lu ; Hong Qin ; Antonio Robles-Kelly ; Chao Zhang ; Ying He
COMMENTS:       submitted for review
HIGHLIGHT:      In this paper, weintroduce a deep patch-based method for 3D human segmentation.

82, TITLE:      An Enhanced Text Classification to Explore Health based Indian Government Policy Tweets
http://arxiv.org/abs/2007.06511
AUTHORS:        Aarzoo Dhiman ; Durga Toshniwal
COMMENTS:       4 pages, 2 figures, 2 tables
HIGHLIGHT:      In this research work, we propose an improved text classification framework that classifies the Twitter data of different health-based government schemes.

83, TITLE:      Evolving Graphical Planner: Contextual Global Planning for Vision-and-Language Navigation
http://arxiv.org/abs/2007.05655
AUTHORS:        Zhiwei Deng ; Karthik Narasimhan ; Olga Russakovsky
HIGHLIGHT:      In this paper, we introduce the Evolving Graphical Planner (EGP), a model that performs global planning for navigation based on raw sensory input.

84, TITLE:      AutoTrajectory: Label-free Trajectory Extraction and Prediction from Videos using Dynamic Points
http://arxiv.org/abs/2007.05719
AUTHORS:        Yuexin Ma ; Xinge ZHU ; Xinjing Cheng ; Ruigang Yang ; Jiming Liu ; Dinesh Manocha
HIGHLIGHT:      In this paper, we present a novel, label-free algorithm, AutoTrajectory, for trajectory extraction and prediction to use raw videos directly.

85, TITLE:      Improved Detection of Adversarial Images Using Deep Neural Networks
http://arxiv.org/abs/2007.05573
AUTHORS:        Yutong Gao ; Yi Pan
HIGHLIGHT:      We propose a new approach called Feature Map Denoising to detect the adversarial inputs and show the performance of detection on the mixed dataset consisting of adversarial examples generated by different attack algorithms, which can be used to associate with any pre-trained DNNs at a low cost.

86, TITLE:      Thirty-seven years of relational Hoare logic: remarks on its principles and history
http://arxiv.org/abs/2007.06421
AUTHORS:        David A. Naumann
COMMENTS:       Submitted to ISOLA 2020
HIGHLIGHT:      Thirty-seven years of relational Hoare logic: remarks on its principles and history

87, TITLE:      Conditional Lower Bound for Inclusion-Based Points-to Analysis
http://arxiv.org/abs/2007.05569
AUTHORS:        Qirun Zhang
HIGHLIGHT:      In this paper, we prove that a truly subcubic $O(n^{3-\delta})$ time combinatorial algorithm for inclusion-based points-to analysis is unlikely: a truly subcubic combinatorial points-to analysis algorithm implies a truly subcubic combinatorial algorithm for Boolean Matrix Multiplication (BMM).

88, TITLE:      A Strong XOR Lemma for Randomized Query Complexity
http://arxiv.org/abs/2007.05580
AUTHORS:        Joshua Brody ; Jae Tak Kim ; Peem Lerdputtipongporn ; Hari Srinivasulu
COMMENTS:       9 pages
HIGHLIGHT:      We give a strong direct sum theorem for computing $xor \circ g$.

89, TITLE:      Multitask Non-Autoregressive Model for Human Motion Prediction
http://arxiv.org/abs/2007.06426
AUTHORS:        Bin Li ; Jian Tian ; Zhongfei Zhang ; Hailin Feng ; Xi Li
HIGHLIGHT:      In this paper, we argue that such issue is mainly caused by adopting autoregressive manner.

90, TITLE:      Vizarel: A System to Help Better Understand RL Agents
http://arxiv.org/abs/2007.05577
AUTHORS:        Shuby Deshpande ; Jeff Schneider
COMMENTS:       Accepted to ICML 2020 Workshop on Human Interpretability in Machine Learning (Spotlight)
HIGHLIGHT:      In this work, we describe our initial attempt at constructing a prototype of these ideas, through identifying possible features that such a system should encapsulate.

91, TITLE:      GGPONC: A Corpus of German Medical Text with Rich Metadata Based on Clinical Practice Guidelines
http://arxiv.org/abs/2007.06400
AUTHORS:        Florian Borchert ; Christina Lohr ; Luise Modersohn ; Thomas Langer ; Markus Follmann ; Jan Philipp Sachs ; Udo Hahn ; Matthieu-P. Schapranow
HIGHLIGHT:      In this work, we present GGPONC (German Guideline Program in Oncology NLP Corpus), a freely distributable German language corpus based on clinical practice guidelines in the field of oncology.

92, TITLE:      Fashion-IQ 2020 Challenge 2nd Place Team's Solution
http://arxiv.org/abs/2007.06404
AUTHORS:        Minchul Shin ; Yoonjae Cho ; Seongwuk Hong
COMMENTS:       4 pages, CVPR 2020 Workshop, Fashion IQ Challenge
HIGHLIGHT:      Given a pair of the image and the text, we present a novel multimodal composition method, RTIC, that can effectively combine the text and the image modalities into a semantic space.

93, TITLE:      Nested Learning For Multi-Granular Tasks
http://arxiv.org/abs/2007.06402
AUTHORS:        Raphaël Achddou ; J. Matias di Martino ; Guillermo Sapiro
HIGHLIGHT:      To address these challenges, we introduce the concept of nested learning: how to obtain a hierarchical representation of the input such that a coarse label can be extracted first, and sequentially refine this representation, if the sample permits, to obtain successively refined predictions, all of them with the corresponding confidence.

94, TITLE:      Deep Cross-Subject Mapping of Neural Activity
http://arxiv.org/abs/2007.06407
AUTHORS:        Marko Angjelichinoski ; Bijan Pesaran ; Vahid Tarokh
COMMENTS:       Submitted to NeurIPS 2020: (15 pages, 7 figures, 5 tables)
HIGHLIGHT:      In this paper, we demonstrate that a neural decoder trained on neural activity signals of one subject can be used to \textit{robustly} decode the motor intentions of a different subject with high reliability.

95, TITLE:      PA-GAN: Progressive Attention Generative Adversarial Network for Facial Attribute Editing
http://arxiv.org/abs/2007.05892
AUTHORS:        Zhenliang He ; Meina Kan ; Jichao Zhang ; Shiguang Shan
COMMENTS:       Code: https://github.com/LynnHo/PA-GAN-Tensorflow
HIGHLIGHT:      To resolve this dilemma, we propose a progressive attention GAN (PA-GAN) for facial attribute editing.

96, TITLE:      HyperGrid: Efficient Multi-Task Transformers with Grid-wise Decomposable Hyper Projections
http://arxiv.org/abs/2007.05891
AUTHORS:        Yi Tay ; Zhe Zhao ; Dara Bahri ; Donald Metzler ; Da-Cheng Juan
HIGHLIGHT:      In this paper, we propose \textsc{HyperGrid}, a new approach for highly effective multi-task learning.

97, TITLE:      Learning Abstract Models for Strategic Exploration and Fast Reward Transfer
http://arxiv.org/abs/2007.05896
AUTHORS:        Evan Zheran Liu ; Ramtin Keramati ; Sudarshan Seshadri ; Kelvin Guu ; Panupong Pasupat ; Emma Brunskill ; Percy Liang
HIGHLIGHT:      Instead, to avoid compounding errors, we propose learning an abstract MDP over abstract states: low-dimensional coarse representations of the state (e.g., capturing agent position, ignoring other objects).

98, TITLE:      Train Your Data Processor: Distribution-Aware and Error-Compensation Coordinate Decoding for Human Pose Estimation
http://arxiv.org/abs/2007.05887
AUTHORS:        Feiyu Yang ; Yu Chen ; Zhe Pan ; Min Zhang ; Min Xue ; Yaoyang Mo ; Yao Zhang ; Guoxiong Guan ; Beibei Qian ; Zhenzhong Xiao ; Zhan Song
COMMENTS:       Improve the state-of-the-art of COCO keypoint detection challenge by 1-2 AP. Project page: https://github.com/fyang235/DAEC
HIGHLIGHT:      Serving as a model-agnostic plug-in, DAEC learns its decoding strategy from training data and remarkably improves the performance of a variety of state-of-the-art human pose estimation models.

99, TITLE:      I3rab: A New Arabic Dependency Treebank Based on Arabic Grammatical Theory
http://arxiv.org/abs/2007.05772
AUTHORS:        Dana Halabi ; Ebaa Fayyoumi ; Arafat Awajan
HIGHLIGHT:      The purpose of this paper is to construct a new Arabic dependency treebank based on the traditional Arabic grammatical theory and the characteristics of the Arabic language, to investigate their effects on the accuracy of statistical parsers.

100, TITLE:     Is Machine Learning Speaking my Language? A Critical Look at the NLP-Pipeline Across 8 Human Languages
http://arxiv.org/abs/2007.05872
AUTHORS:        Esma Wali ; Yan Chen ; Christopher Mahoney ; Thomas Middleton ; Marzieh Babaeianjelodar ; Mariama Njie ; Jeanna Neefe Matthews
COMMENTS:       Participatory Approaches to Machine Learning Workshop, 37th International Conference on Machine Learning
HIGHLIGHT:      In this paper, a team including speakers of 8 languages - English, Chinese, Urdu, Farsi, Arabic, French, Spanish, and Wolof - takes a critical look at the typical NLP pipeline and how even when a language is technically supported, substantial caveats remain to prevent full participation.

101, TITLE:     Applying recent advances in Visual Question Answering to Record Linkage
http://arxiv.org/abs/2007.05881
AUTHORS:        Marko Smilevski
COMMENTS:       48 pages, 15 figures, 6 tables
HIGHLIGHT:      This field has not been explored in research and we propose two solutions based on Deep Learning architectures that are inspired by recent work in Visual Question Answering.

102, TITLE:     Leaky Integrate-and-Fire Spiking Neuron with Learnable Membrane Time Parameter
http://arxiv.org/abs/2007.05785
AUTHORS:        Wei Fang
HIGHLIGHT:      In this article, we propose a novel spiking neuron, namely parametric Leaky Integrate-and-Fire (PLIF) neuron, whose $\tau$ is a learnable parameter rather than an empirical hyper-parameter.

103, TITLE:     A Hybrid Multi-Objective Carpool Route Optimization Technique using Genetic Algorithm and A* Algorithm
http://arxiv.org/abs/2007.05781
AUTHORS:        Romit S Beed ; Sunita Sarkar ; Arindam Roy ; Suvranil D Biswas ; Suhana Biswas
HIGHLIGHT:      This work presents a hybrid GA-A* algorithm to obtain optimal routes for the carpooling problem in the domain of multi-objective optimization having multiple conflicting objectives.

104, TITLE:     Pyramid Scale Network for Crowd Counting
http://arxiv.org/abs/2007.05779
AUTHORS:        Junhao Cheng ; Zhuojun Chen ; XinYu Zhang ; Yizhou Li ; Xiaoyuan Jing
HIGHLIGHT:      In this paper, we propose a novel crowd counting framework called Pyramid Scale Network (PSNet) to explicitly address these issues.

105, TITLE:     Driver Behavior Modelling at the Urban Intersection via Canonical Correlation Analysis
http://arxiv.org/abs/2007.05751
AUTHORS:        Zirui Li ; Chao Lu ; Cheng Gong ; Cheng Gong ; Jinghang Li ; Lianzhen Wei
COMMENTS:       2020 3rd IEEE International Conference on Unmanned Systems (ICUS)
HIGHLIGHT:      In this research, a canonical correlation analysis (CCA)-based framework is proposed.

106, TITLE:     Neuromorphic Processing and Sensing: Evolutionary Progression of AI to Spiking
http://arxiv.org/abs/2007.05606
AUTHORS:        Philippe Reiter ; Geet Rose Jose ; Spyridon Bizmpikis ; Ionela-Ancuţa Cîrjilă
COMMENTS:       15 pages, 13 figures
HIGHLIGHT:      This paper explains the theoretical workings of neuromorphic technologies based on spikes, and overviews the state-of-art in hardware processors, software platforms and neuromorphic sensing devices.

107, TITLE:     Class LM and word mapping for contextual biasing in End-to-End ASR
http://arxiv.org/abs/2007.05609
AUTHORS:        Rongqing Huang ; Ossama Abdel-hamid ; Xinwei Li ; Gunnar Evermann
HIGHLIGHT:      In this paper, we propose to train a context aware E2E model and allow the beam search to traverse into the context FST during inference.

108, TITLE:     Image Captioning with Compositional Neural Module Networks
http://arxiv.org/abs/2007.05608
AUTHORS:        Junjiao Tian ; Jean Oh
COMMENTS:       International Joint Conference on Artificial Intelligence (IJCAI-19)
HIGHLIGHT:      Inspired by the idea of the compositional neural module networks in the visual question answering task, we introduce a hierarchical framework for image captioning that explores both compositionality and sequentiality of natural language.

109, TITLE:     Relation-Guided Representation Learning
http://arxiv.org/abs/2007.05742
AUTHORS:        Zhao Kang ; Xiao Lu ; Jian Liang ; Kun Bai ; Zenglin Xu
COMMENTS:       Appear in Neural Networks
HIGHLIGHT:      In this work, we propose a new representation learning method that explicitly models and leverages sample relations, which in turn is used as supervision to guide the representation learning.

110, TITLE:     Distangling Biological Noise in Cellular Images with a focus on Explainability
http://arxiv.org/abs/2007.05743
AUTHORS:        Manik Sharma ; Ganapathy Krishnamurthi
COMMENTS:       13 Pages, 12 figures
HIGHLIGHT:      This work aims at solving a part of this problem by creating a cellular image classification model which can decipher the genetic perturbations in cell (occurring naturally or artificially).

111, TITLE:     Weighted First-Order Model Counting in the Two-Variable Fragment With Counting Quantifiers
http://arxiv.org/abs/2007.05619
AUTHORS:        Ondrej Kuzelka
HIGHLIGHT:      In this paper we extend this result to the two-variable fragment with counting quantifiers.

112, TITLE:     GloVeInit at SemEval-2020 Task 1: Using GloVe Vector Initialization for Unsupervised Lexical Semantic Change Detection
http://arxiv.org/abs/2007.05618
AUTHORS:        Vaibhav Jain
COMMENTS:       To be presented at the 2020 International Workshop on Semantic Evaluation
HIGHLIGHT:      This paper presents a vector initialization approach for the SemEval2020 Task 1: Unsupervised Lexical Semantic Change Detection.

113, TITLE:     Multi-Dialect Arabic BERT for Country-Level Dialect Identification
http://arxiv.org/abs/2007.05612
AUTHORS:        Bashar Talafha ; Mohammad Ali ; Muhy Eddin Za'ter ; Haitham Seelawi ; Ibraheem Tuffaha ; Mostafa Samir ; Wael Farhan ; Hussein T. Al-Natsheh
COMMENTS:       Accepted at the Fifth Arabic Natural Language Processing Workshop (WANLP2020) co-located with the 28th International Conference on Computational Linguistics (COLING'2020), Barcelona, Spain, 12 Dec. 2020
HIGHLIGHT:      In this paper, we present the experiments conducted, and the models developed by our competing team, Mawdoo3 AI, along the way to achieving our winning solution to subtask 1 of the Nuanced Arabic Dialect Identification (NADI) shared task.

114, TITLE:     Deep Contextual Clinical Prediction with Reverse Distillation
http://arxiv.org/abs/2007.05611
AUTHORS:        Rohan S. Kodialam ; Rebecca Boiarsky ; David Sontag
HIGHLIGHT:      In this work, motivated by the task of clinical prediction from insurance claims, we present a new technique called reverse distillation which pretrains deep models by using high-performing linear models for initialization.

115, TITLE:     Batch-Incremental Triplet Sampling for Training Triplet Networks Using Bayesian Updating Theorem
http://arxiv.org/abs/2007.05610
AUTHORS:        Milad Sikaroudi ; Benyamin Ghojogh ; Fakhri Karray ; Mark Crowley ; H. R. Tizhoosh
COMMENTS:       The first two authors contributed equally to this work
HIGHLIGHT:      In this work, we sample triplets from distributions of data rather than from existing instances.

116, TITLE:     Quantization in Relative Gradient Angle Domain For Building Polygon Estimation
http://arxiv.org/abs/2007.05617
AUTHORS:        Yuhao Chen ; Yifan Wu ; Linlin Xu ; Alexander Wong
HIGHLIGHT:      In this paper, we leverage the performance of CNNs, and propose a module that uses prior knowledge of building corners to create angular and concise building polygons from CNN segmentation outputs.

117, TITLE:     PCAMs: Weakly Supervised Semantic Segmentation Using Point Supervision
http://arxiv.org/abs/2007.05615
AUTHORS:        R. Austin McEver ; B. S. Manjunath
HIGHLIGHT:      This paper presents a novel procedure for producing semantic segmentation from images given some point level annotations.

118, TITLE:     Generative Graph Perturbations for Scene Graph Prediction
http://arxiv.org/abs/2007.05756
AUTHORS:        Boris Knyazev ; Harm de Vries ; Cătălina Cangea ; Graham W. Taylor ; Aaron Courville ; Eugene Belilovsky
COMMENTS:       https://oolworkshop.github.io/program/ool_21.html, ICML Workshop 2020 on "Object-Oriented Learning (OOL): Perception, Representation, and Reasoning"
HIGHLIGHT:      To increase their diversity, we propose several strategies to perturb the conditioning.

119, TITLE:     Probability Learning based Tabu Search for the Budgeted Maximum Coverage Problem
http://arxiv.org/abs/2007.05971
AUTHORS:        Liwen Li ; Zequn Wei ; Jin-Kao Hao ; Kun He
HIGHLIGHT:      In this work, we deal with the Budgeted Maximum Coverage Problem (BMCP), which is a generalized 0-1 knapsack problem.

120, TITLE:     Towards practical lipreading with distilled and efficient models
http://arxiv.org/abs/2007.06504
AUTHORS:        Pingchuan Ma ; Brais Martinez ; Stavros Petridis ; Maja Pantic
HIGHLIGHT:      In this work, we propose a series of innovations that significantly bridge that gap: first, we raise the state-of-the-art performance by a wide margin on LRW and LRW-1000 to 88.6% and 46.6%, respectively, through careful optimization.

121, TITLE:     IllumiNet: Transferring Illumination from Planar Surfaces to Virtual Objects in Augmented Reality
http://arxiv.org/abs/2007.05981
AUTHORS:        Di Xu ; Zhen Li ; Yanning Zhang ; Qi Cao
HIGHLIGHT:      This paper presents an illumination estimation method for virtual objects in real environment by learning.

122, TITLE:     Stance Detection in Web and Social Media: A Comparative Study
http://arxiv.org/abs/2007.05976
AUTHORS:        Shalmoli Ghosh ; Prajwal Singhania ; Siddharth Singh ; Koustav Rudra ; Saptarshi Ghosh
HIGHLIGHT:      In this work, we explore the reproducibility of several existing stance detection models, including both neural models and classical classifier-based models.

123, TITLE:     Self-Supervised Drivable Area and Road Anomaly Segmentation using RGB-D Data for Robotic Wheelchairs
http://arxiv.org/abs/2007.05950
AUTHORS:        Hengli Wang ; Yuxiang Sun ; Ming Liu
COMMENTS:       Published in IEEE Robotics and Automation Letters (RA-L); 8 pages, 8 figures and 3 tables
HIGHLIGHT:      We contribute to the solution of this problem for the task of drivable area and road anomaly segmentation by proposing a self-supervised learning approach.

124, TITLE:     Anomaly Detection-Based Unknown Face Presentation Attack Detection
http://arxiv.org/abs/2007.05856
AUTHORS:        Yashasvi Baweja ; Poojan Oza ; Pramuditha Perera ; Vishal M. Patel
HIGHLIGHT:      In this paper, we present a deep-learning solution for anomaly detection-based spoof attack detection where both classifier and feature representations are learned together end-to-end.

125, TITLE:     Complex Wavelet SSIM based Image Data Augmentation
http://arxiv.org/abs/2007.05853
AUTHORS:        Ritin Raveendran ; Aviral Singh ; Rajesh Kumar M
HIGHLIGHT:      In this paper, we look particularly at the MNIST handwritten dataset an image dataset used for digit recognition, and the methods of data augmentation done on this data set.

126, TITLE:     Efficient resource management in UAVs for Visual Assistance
http://arxiv.org/abs/2007.05854
AUTHORS:        Bapireddy Karri ; Sudip Misra ; Senior Member IEEE
COMMENTS:       7 pages
HIGHLIGHT:      This projects describes a novel method to optimize the general image processing tasks like object tracking and object detection for UAV hardware in real time scenarios without affecting the flight time and not tampering the latency and accuracy of these models.

127, TITLE:     Understanding Object Detection Through An Adversarial Lens
http://arxiv.org/abs/2007.05828
AUTHORS:        Ka-Ho Chow ; Ling Liu ; Mehmet Emre Gursoy ; Stacey Truex ; Wenqi Wei ; Yanzhao Wu
HIGHLIGHT:      This paper presents a framework for analyzing and evaluating vulnerabilities of the state-of-the-art object detectors under an adversarial lens, aiming to analyze and demystify the attack strategies, adverse effects, and costs, as well as the cross-model and cross-resolution transferability of attacks.

128, TITLE:     Representation Learning via Adversarially-Contrastive Optimal Transport
http://arxiv.org/abs/2007.05840
AUTHORS:        Anoop Cherian ; Shuchin Aeron
COMMENTS:       Accepted at ICML 2020
HIGHLIGHT:      In this paper, we study the problem of learning compact (low-dimensional) representations for sequential data that captures its implicit spatio-temporal cues.

129, TITLE:     Control as Hybrid Inference
http://arxiv.org/abs/2007.05838
AUTHORS:        Alexander Tschantz ; Beren Millidge ; Anil K. Seth ; Christopher L. Buckley
HIGHLIGHT:      We present an implementation of CHI which naturally mediates the balance between iterative and amortised inference.

130, TITLE:     Deep Network Interpolation for Accelerated Parallel MR Image Reconstruction
http://arxiv.org/abs/2007.05993
AUTHORS:        Chen Qin ; Jo Schlemper ; Kerstin Hammernik ; Jinming Duan ; Ronald M Summers ; Daniel Rueckert
COMMENTS:       Presented at 2020 ISMRM Conference & Exhibition (Abstract #4958)
HIGHLIGHT:      We present a deep network interpolation strategy for accelerated parallel MR image reconstruction.

131, TITLE:     Differentiable Programming for Hyperspectral Unmixing using a Physics-based Dispersion Model
http://arxiv.org/abs/2007.05996
AUTHORS:        John Janiczek ; Parth Thaker ; Gautam Dasarathy ; Christopher S. Edwards ; Philip Christensen ; Suren Jayasuriya
COMMENTS:       36 pages, 11 figures. Accepted to European Conference on Computer Vision (ECCV) 2020
HIGHLIGHT:      In this paper, spectral variation is considered from a physics-based approach and incorporated into an end-to-end spectral unmixing algorithm via differentiable programming.

132, TITLE:     Dual Adversarial Network: Toward Real-world Noise Removal and Noise Generation
http://arxiv.org/abs/2007.05946
AUTHORS:        Zongsheng Yue ; Qian Zhao ; Lei Zhang ; Deyu Meng
COMMENTS:       Accepted by ECCV 2020
HIGHLIGHT:      In this work, we propose a novel unified framework to simultaneously deal with the noise removal and noise generation tasks.

133, TITLE:     Fruit classification using deep feature maps in the presence of deceptive similar classes
http://arxiv.org/abs/2007.05942
AUTHORS:        Mohit Dandekar ; Narinder Singh Punn ; Sanjay Kumar Sonbhadra ; Sonali Agarwal
HIGHLIGHT:      The objective of the present research is to address the challenge of classification of deceptively similar multi-granular objects with an ensemble approach thfat utilizes activations from multiple layers of CNN (deep features).

134, TITLE:     Relational-Grid-World: A Novel Relational Reasoning Environment and An Agent Model for Relational Information Extraction
http://arxiv.org/abs/2007.05961
AUTHORS:        Faruk Kucuksubasi ; Elif Surer
HIGHLIGHT:      In this study, we present a model-free RL architecture that is supported with explicit relational representations of the environmental objects.

135, TITLE:     Abstract Universal Approximation for Neural Networks
http://arxiv.org/abs/2007.06093
AUTHORS:        Zi Wang ; Aws Albarghouthi ; Somesh Jha
HIGHLIGHT:      We present a theoretical result that demonstrates the power of numerical domains, namely, the simple interval domain, for analysis of neural networks.

136, TITLE:     Visualizing Classification Structure in Deep Neural Networks
http://arxiv.org/abs/2007.06068
AUTHORS:        Bilal Alsallakh ; Zhixin Yan ; Shabnam Ghaffarzadegan ; Zeng Dai ; Liu Ren
COMMENTS:       2020 ICML Workshop on Human Interpretability in Machine Learning (WHI 2020)
HIGHLIGHT:      We propose a measure to compute class similarity in large-scale classification based on prediction scores.

137, TITLE:     Adversarial Self-Supervised Learning for Semi-Supervised 3D Action Recognition
http://arxiv.org/abs/2007.05934
AUTHORS:        Chenyang Si ; Xuecheng Nie ; Wei Wang ; Liang Wang ; Tieniu Tan ; Jiashi Feng
COMMENTS:       Accepted by ECCV2020
HIGHLIGHT:      To address these issues, we present Adversarial Self-Supervised Learning (ASSL), a novel framework that tightly couples SSL and the semi-supervised scheme via neighbor relation exploration and adversarial learning.

138, TITLE:     Exploiting Uncertainties from Ensemble Learners to Improve Decision-Making in Healthcare AI
http://arxiv.org/abs/2007.06063
AUTHORS:        Yingshui Tan ; Baihong Jin ; Xiangyu Yue ; Yuxin Chen ; Alberto Sangiovanni Vincentelli
COMMENTS:       Preprint of submission to NeurIPS 2020
HIGHLIGHT:      In this paper, we study the following key research question in the selection of uncertainty metrics: when does an uncertainty metric outperforms another?

139, TITLE:     Pose-aware Adversarial Domain Adaptation for Personalized Facial Expression Recognition
http://arxiv.org/abs/2007.05932
AUTHORS:        Guang Liang ; Shangfei Wang ; Can Wang
HIGHLIGHT:      In this paper, we propose a novel unsupervised adversarial domain adaptation method which can alleviate both variations at the same time.

140, TITLE:     Fine-grained Language Identification with Multilingual CapsNet Model
http://arxiv.org/abs/2007.06078
AUTHORS:        Mudit Verma ; Arun Balaji Buduru
COMMENTS:       5 pages, 6 figures
HIGHLIGHT:      Hence in this work, a real-time language detection approach to detect spoken language from 5 seconds' audio clips with an accuracy of 91.8\% is presented with exiguous data requirements and minimal pre-processing.

141, TITLE:     Framework for Passenger Seat Availability Using Face Detection in Passenger Bus
http://arxiv.org/abs/2007.05906
AUTHORS:        Khawar Islam ; Uzma Afzal
HIGHLIGHT:      We propose a Face Detection based Framework (FDF) to determine passenger seat availability in a camera-equipped bus through face detection which is based on background subtraction to count empty, filled, and total seats.

142, TITLE:     Sparse Graph to Sequence Learning for Vision Conditioned Long Textual Sequence Generation
http://arxiv.org/abs/2007.06077
AUTHORS:        Aditya Mogadala ; Marius Mosbach ; Dietrich Klakow
COMMENTS:       International Conference on Machine Learning (ICML) 2020 Workshop (https://logicalreasoninggnn.github.io/)
HIGHLIGHT:      In this paper, we mask this Vision-to-Sequence as Graph-to-Sequence learning problem and approach it with the Transformer architecture.

143, TITLE:     Editable AI: Mixed Human-AI Authoring of Code Patterns
http://arxiv.org/abs/2007.05902
AUTHORS:        Kartik Chugh ; Andrea Y. Solis ; Thomas D. LaToza
HIGHLIGHT:      To surface these patterns to developers and support developers in authoring consistent with these patterns, we propose a mixed human-AI technique for creating code patterns.

144, TITLE:     Lightweight Modules for Efficient Deep Learning based Image Restoration
http://arxiv.org/abs/2007.05835
AUTHORS:        Avisek Lahiri ; Sourav Bairagya ; Sutanu Bera ; Siddhant Haldar ; Prabir Kumar Biswas
COMMENTS:       Accepted at: IEEE Transactions on Circuits and Systems for Video Technology (Early Access Print) | |Codes Available at: https://github.com/avisekiit/TCSVT-LightWeight-CNNs | Supplementary Document at: https://drive.google.com/file/d/1BQhkh33Sen-d0qOrjq5h8ahw2VCUIVLg/view?usp=sharing
HIGHLIGHT:      In this paper, we propose several lightweight low-level modules which can be used to create a computationally low cost variant of a given baseline model.

145, TITLE:     Meta Soft Label Generation for Noisy Labels
http://arxiv.org/abs/2007.05836
AUTHORS:        Görkem Algan ; Ilkay Ulusoy
HIGHLIGHT:      To address this problem, we propose a Meta Soft Label Generation algorithm called MSLG, which can jointly generate soft labels using meta-learning techniques and learn DNN parameters in an end-to-end fashion.

146, TITLE:     AutoEmbedder: A semi-supervised DNN embedding system for clustering
http://arxiv.org/abs/2007.05830
AUTHORS:        Abu Quwsar Ohi ; M. F. Mridha ; Farisa Benta Safir ; Md. Abdul Hamid ; Muhammad Mostafa Monowar
COMMENTS:       The manuscript is accepted and published in Knowledge-Based System
HIGHLIGHT:      This paper introduces a novel embedding system named AutoEmbedder, that downsamples higher dimensional data to clusterable embedding points.

147, TITLE:     The ASRU 2019 Mandarin-English Code-Switching Speech Recognition Challenge: Open Datasets, Tracks, Methods and Results
http://arxiv.org/abs/2007.05916
AUTHORS:        Xian Shi ; Qiangze Feng ; Lei Xie
HIGHLIGHT:      This paper describes the design and main outcomes of the ASRU 2019 Mandarin-English code-switching speech recognition challenge, which aims to improve the ASR performance in Mandarin-English code-switching situation.

148, TITLE:     Two-Stream Deep Feature Modelling for Automated Video Endoscopy Data Analysis
http://arxiv.org/abs/2007.05914
AUTHORS:        Harshala Gammulle ; Simon Denman ; Sridha Sridharan ; Clinton Fookes
COMMENTS:       Accepted for Publication at MICCAI 2020
HIGHLIGHT:      To further the development of such methods, we propose a two-stream model for endoscopic image analysis.

149, TITLE:     Understanding Adversarial Examples from the Mutual Influence of Images and Perturbations
http://arxiv.org/abs/2007.06189
AUTHORS:        Chaoning Zhang ; Philipp Benz ; Tooba Imtiaz ; In-So Kweon
COMMENTS:       Accepted at CVPR 2020
HIGHLIGHT:      We propose to treat the DNN logits as a vector for feature representation, and exploit them to analyze the mutual influence of two independent inputs based on the Pearson correlation coefficient (PCC).

150, TITLE:     Learning to Learn Parameterized Classification Networks for Scalable Input Images
http://arxiv.org/abs/2007.06181
AUTHORS:        Duo Li ; Anbang Yao ; Qifeng Chen
COMMENTS:       Accepted by ECCV 2020. Code and models are available at https://github.com/d-li14/SAN
HIGHLIGHT:      To achieve efficient and flexible image classification at runtime, we employ meta learners to generate convolutional weights of main networks for various input scales and maintain privatized Batch Normalization layers per scale.

151, TITLE:     Fusing Motion Patterns and Key Visual Information for Semantic Event Recognition in Basketball Videos
http://arxiv.org/abs/2007.06288
AUTHORS:        Lifang Wu ; Zhou Yang ; Qi Wang ; Meng Jian ; Boxuan Zhao ; Junchi Yan ; Chang Wen Chen
HIGHLIGHT:      Based on the observations, we propose a scheme to fuse global and local motion patterns (MPs) and key visual information (KVI) for semantic event recognition in basketball videos.

152, TITLE:     Accelerated FBP for computed tomography image reconstruction
http://arxiv.org/abs/2007.06289
AUTHORS:        Anastasiya Dolmatova ; Marina Chukalina ; Dmitry Nikolaev
HIGHLIGHT:      In this paper, we propose a novel approach that reduces the computational complexity of the algorithm to $\Theta(N^2\log N)$ addition operations avoiding Fourier space.

153, TITLE:     Data from Model: Extracting Data from Non-robust and Robust Models
http://arxiv.org/abs/2007.06196
AUTHORS:        Philipp Benz ; Chaoning Zhang ; Tooba Imtiaz ; In-So Kweon
COMMENTS:       Accepted at the CVPR 2020 Workshop on Adversarial Machine Learning in Computer Vision
HIGHLIGHT:      We repeat the process of Data to Model (DtM) and Data from Model (DfM) in sequence and explore the loss of feature mapping information by measuring the accuracy drop on the original validation dataset.

154, TITLE:     Beyond Graph Neural Networks with Lifted Relational Neural Networks
http://arxiv.org/abs/2007.06286
AUTHORS:        Gustav Sourek ; Filip Zelezny ; Ondrej Kuzelka
COMMENTS:       Submitted to MLJ's Special Track on Learning and Reasoning (May 15th 2020 cut-off) http://lr2020.iit.demokritos.gr/
HIGHLIGHT:      We demonstrate a declarative differentiable programming framework based on the language of Lifted Relational Neural Networks, where small parameterized logic programs are used to encode relational learning scenarios.

155, TITLE:     Reducing Language Biases in Visual Question Answering with Visually-Grounded Question Encoder
http://arxiv.org/abs/2007.06198
AUTHORS:        Gouthaman KV ; Anurag Mittal
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this work, we propose a novel model-agnostic question encoder, Visually-Grounded Question Encoder (VGQE), for VQA that reduces this effect.

156, TITLE:     CheXphoto: 10,000+ Smartphone Photos and Synthetic Photographic Transformations of Chest X-rays for Benchmarking Deep Learning Robustness
http://arxiv.org/abs/2007.06199
AUTHORS:        Nick A. Phillips ; Pranav Rajpurkar ; Mark Sabini ; Rayan Krishnan ; Sharon Zhou ; Anuj Pareek ; Nguyet Minh Phu ; Chris Wang ; Andrew Y. Ng ; Matthew P. Lungren
HIGHLIGHT:      We introduce CheXphoto, a dataset of smartphone photos and synthetic photographic transformations of chest x-rays sampled from the CheXpert dataset. We release this dataset as a resource for testing and improving the robustness of deep learning algorithms for automated chest x-ray interpretation on smartphone photos of chest x-rays.

157, TITLE:     Probabilistic bounds on data sensitivity in deep rectifier networks
http://arxiv.org/abs/2007.06192
AUTHORS:        Blaine Rister ; Daniel L. Rubin
HIGHLIGHT:      In this work, we provide a simple and rigorous proof of that result.

158, TITLE:     PSConv: Squeezing Feature Pyramid into One Compact Poly-Scale Convolutional Layer
http://arxiv.org/abs/2007.06191
AUTHORS:        Duo Li ; Anbang Yao ; Qifeng Chen
COMMENTS:       Accepted by ECCV 2020. Code and models are available at https://github.com/d-li14/PSConv
HIGHLIGHT:      We bridge this regret by exploiting multi-scale features in a finer granularity.

159, TITLE:     Low to High Dimensional Modality Hallucination using Aggregated Fields of View
http://arxiv.org/abs/2007.06166
AUTHORS:        Kausic Gunasekar ; Qiang Qiu ; Yezhou Yang
HIGHLIGHT:      We present a novel hallucination architecture that aggregates information from multiple fields of view of the local neighborhood to recover the lost information from the extant modality.

160, TITLE:     Do You Have the Right Scissors? Tailoring Pre-trained Language Models via Monte-Carlo Methods
http://arxiv.org/abs/2007.06162
AUTHORS:        Ning Miao ; Yuxuan Song ; Hao Zhou ; Lei Li
COMMENTS:       Accepted by ACL 2020
HIGHLIGHT:      In this paper, we propose MC-Tailor, a novel method to alleviate the above issue in text generation tasks by truncating and transferring the probability mass from over-estimated regions to under-estimated ones.

161, TITLE:     Bridging Maximum Likelihood and Adversarial Learning via $α$-Divergence
http://arxiv.org/abs/2007.06178
AUTHORS:        Miaoyun Zhao ; Yulai Cong ; Shuyang Dai ; Lawrence Carin
COMMENTS:       AAAI 2020
HIGHLIGHT:      We reveal that generalizations of the $\alpha$-Bridge are closely related to approaches developed recently to regularize adversarial learning, providing insights into that prior work, and further understanding of why the $\alpha$-Bridge performs well in practice.

162, TITLE:     Generating Fluent Adversarial Examples for Natural Languages
http://arxiv.org/abs/2007.06174
AUTHORS:        Huangzhao Zhang ; Hao Zhou ; Ning Miao ; Lei Li
COMMENTS:       Accepted by ACL 2019
HIGHLIGHT:      In this paper, we propose MHA, which addresses both problems by performing Metropolis-Hastings sampling, whose proposal is designed with the guidance of gradients.

163, TITLE:     Coarse scale representation of spiking neural networks: backpropagation through spikes and application to neuromorphic hardware
http://arxiv.org/abs/2007.06176
AUTHORS:        Angel Yanguas-Gil
COMMENTS:       Paper accepted in ICONS 2020
HIGHLIGHT:      In this work we explore recurrent representations of leaky integrate and fire neurons operating at a timescale equal to their absolute refractory period.

164, TITLE:     Fine-Grained Crowd Counting
http://arxiv.org/abs/2007.06146
AUTHORS:        Jia Wan ; Nikil Senthil Kumar ; Antoni B. Chan
HIGHLIGHT:      In this paper, we propose fine-grained crowd counting, which differentiates a crowd into categories based on the low-level behavior attributes of the individuals (e.g. standing/sitting or violent behavior) and then counts the number of people in each category. To enable research in this area, we construct a new dataset of four real-world fine-grained counting tasks: traveling direction on a sidewalk, standing or sitting, waiting in line or not, and exhibiting violent behavior or not.

165, TITLE:     Gender Classification and Bias Mitigation in Facial Images
http://arxiv.org/abs/2007.06141
AUTHORS:        Wenying Wu ; Pavlos Protopapas ; Zheng Yang ; Panagiotis Michalatos
COMMENTS:       9 pages
HIGHLIGHT:      In this paper, we began by conducting surveys on existing benchmark databases for facial recognition and gender classification tasks.

166, TITLE:     Embedded Deep Bilinear Interactive Information and Selective Fusion for Multi-view Learning
http://arxiv.org/abs/2007.06143
AUTHORS:        Jinglin Xu ; Wenbin Li ; Jiantao Shen ; Xinwang Liu ; Peicheng Zhou ; Xiangsen Zhang ; Xiwen Yao ; Junwei Han
HIGHLIGHT:      To fulfill this goal, we propose a novel multi-view learning framework to make the multi-view classification better aimed at the above-mentioned two aspects.

167, TITLE:     Temporal Self-Ensembling Teacher for Semi-Supervised Object Detection
http://arxiv.org/abs/2007.06144
AUTHORS:        Cong Chen ; Shouyang Dong ; Ye Tian abd Kunlin Cao ; Li Liu ; Yuanhao Guo
COMMENTS:       11 papges, 3 figures, preprint for submission
HIGHLIGHT:      To solve these problems, we propose the Temporal Self-Ensembling Teacher (TSE-T) model on top of the KD framework.

168, TITLE:     Deep Reinforced Attention Learning for Quality-Aware Visual Recognition
http://arxiv.org/abs/2007.06156
AUTHORS:        Duo Li ; Qifeng Chen
COMMENTS:       Accepted by ECCV 2020
HIGHLIGHT:      In this paper, we build upon the weakly-supervised generation mechanism of intermediate attention maps in any convolutional neural networks and disclose the effectiveness of attention modules more straightforwardly to fully exploit their potential.

169, TITLE:     AI Playground: Unreal Engine-based Data Ablation Tool for Deep Learning
http://arxiv.org/abs/2007.06153
AUTHORS:        Mehdi Mousavi ; Aashis Khanal ; Rolando Estrada
COMMENTS:       14 pages, 7 figures
HIGHLIGHT:      In this paper, we present AI Playground (AIP), an open-source, Unreal Engine-based tool for generating and labeling virtual image data. To validate our proposed tool, we generated eight datasets of otherwise identical but varying lighting and fidelity conditions.

170, TITLE:     MS-NAS: Multi-Scale Neural Architecture Search for Medical Image Segmentation
http://arxiv.org/abs/2007.06151
AUTHORS:        Xingang Yan ; Weiwen Jiang ; Yiyu Shi ; Cheng Zhuo
HIGHLIGHT:      This paper presents a Multi-Scale NAS (MS-NAS) framework that is featured with multi-scale search space from network backbone to cell operation, and multi-scale fusion capability to fuse features with different sizes.

171, TITLE:     Universal-to-Specific Framework for Complex Action Recognition
http://arxiv.org/abs/2007.06149
AUTHORS:        Peisen Zhao ; Lingxi Xie ; Ya Zhang ; Qi Tian
COMMENTS:       13 pages, 8 figures
HIGHLIGHT:      Inspired by a common flowchart based on the human decision-making process that first narrows down the probable classes and then applies a "rethinking" process for finer-level recognition, we propose an effective universal-to-specific (U2S) framework for complex action recognition.

172, TITLE:     EAGLE: Large-scale Dataset for Vehicle Detection in Aerial Imagery
http://arxiv.org/abs/2007.06124
AUTHORS:        Seyed Majid Azimi ; Reza Bahmanyar ; Corenin Henry ; Franz Kurz
HIGHLIGHT:      To address this issue, we introduce EAGLE (oriEnted object detection using Aerial imaGery in real-worLd scEnarios), a large-scale dataset for multi-class vehicle detection with object orientation information in aerial imagery.

173, TITLE:     A Comparative Study on Polyp Classification using Convolutional Neural Networks
http://arxiv.org/abs/2007.06071
AUTHORS:        Krushi Patel ; Kaidong Li ; Ke Tao ; Quan Wang ; Ajay Bansal ; Amit Rastogi ; Guanghui Wang
HIGHLIGHT:      In this work, we compare the performance of the state-of-the-art general object classification models for polyp classification.

174, TITLE:     Learning to associate detections for real-time multiple object tracking
http://arxiv.org/abs/2007.06041
AUTHORS:        Michel Meneses ; Leonardo Matos ; Bruno Prado ; André de Carvalho ; Hendrik Macedo
COMMENTS:       8 pages, 6 figures
HIGHLIGHT:      In this study, it is investigated the use of artificial neural networks to learning a similarity function that can be used among detections.

175, TITLE:     It Is Likely That Your Loss Should be a Likelihood
http://arxiv.org/abs/2007.06059
AUTHORS:        Mark Hamilton ; Evan Shelhamer ; William T. Freeman
HIGHLIGHT:      We survey and systematically evaluate how to parameterize and apply likelihood parameters for robust modeling and re-calibration.

176, TITLE:     The Impossibility Theorem of Machine Fairness -- A Causal Perspective
http://arxiv.org/abs/2007.06024
AUTHORS:        Kailash Karthik S
HIGHLIGHT:      In this report, causal perspective to the impossibility theorem of fairness is presented along with a causal goal for machine fairness.

177, TITLE:     Improving Maximum Likelihood Training for Text Generation with Density Ratio Estimation
http://arxiv.org/abs/2007.06018
AUTHORS:        Yuxuan Song ; Ning Miao ; Hao Zhou ; Lantao Yu ; Mingxuan Wang ; Lei Li
COMMENTS:       Accepted to International Conference on Artificial Intelligence and Statistics 2020
HIGHLIGHT:      In this paper, we propose{\psi}-MLE, a new training scheme for auto-regressive sequence generative models, which is effective and stable when operating at large sample space encountered in text generation.



==========Updates to Previous Papers==========
1, TITLE:       REMIND Your Neural Network to Prevent Catastrophic Forgetting
http://arxiv.org/abs/1910.02509
AUTHORS:        Tyler L. Hayes ; Kushal Kafle ; Robik Shrestha ; Manoj Acharya ; Christopher Kanan
COMMENTS:       To appear in the European Conference on Computer Vision (ECCV-2020)
HIGHLIGHT:      Here, we propose REMIND, a brain-inspired approach that enables efficient replay with compressed representations.

2, TITLE:       Syn-QG: Syntactic and Shallow Semantic Rules for Question Generation
http://arxiv.org/abs/2004.08694
AUTHORS:        Kaustubh D. Dhole ; Christopher D. Manning
COMMENTS:       In Proceedings of the 2020 Annual Conference of the Association for Computational Linguistics (ACL 2020)
HIGHLIGHT:      We implement this observation by developing Syn-QG, a set of transparent syntactic rules leveraging universal dependencies, shallow semantic parsing, lexical resources, and custom rules which transform declarative sentences into question-answer pairs.

3, TITLE:       Deep Decomposition Learning for Inverse Imaging Problems
http://arxiv.org/abs/1911.11028
AUTHORS:        Dongdong Chen ; Mike E. Davies
COMMENTS:       To appear in ECCV 2020
HIGHLIGHT:      In this paper, inspired by the geometry that data can be decomposed by two components from the null-space of the forward operator and the range space of its pseudo-inverse, we train neural networks to learn the two components and therefore learn the decomposition, \ieB we explicitly reformulate the neural network layers as learning range-nullspace decomposition functions with reference to the layer inputs, instead of learning unreferenced functions.

4, TITLE:       Domain Adaptation without Source Data
http://arxiv.org/abs/2007.01524
AUTHORS:        Youngeun Kim ; Sungeun Hong ; Donghyeon Cho ; Hyoungseob Park ; Priyadarshini Panda
COMMENTS:       13 pages
HIGHLIGHT:      To avoid accessing source data which may contain sensitive information, we introduce source data-free domain adaptation (SFDA).

5, TITLE:       A Survey on Sensor Technologies for Unmanned Ground Vehicles
http://arxiv.org/abs/2007.01992
AUTHORS:        Qi Liu ; Shihua Yuan ; Zirui Li
COMMENTS:       9 pages,6 tables
HIGHLIGHT:      This paper proposes a brief review on sensor technologies for UGVs.

6, TITLE:       Point-Set Anchors for Object Detection, Instance Segmentation and Pose Estimation
http://arxiv.org/abs/2007.02846
AUTHORS:        Fangyun Wei ; Xiao Sun ; Hongyang Li ; Jingdong Wang ; Stephen Lin
COMMENTS:       To appear in ECCV 2020
HIGHLIGHT:      To facilitate inference, we propose to instead perform regression from a set of points placed at more advantageous positions.

7, TITLE:       X-ModalNet: A Semi-Supervised Deep Cross-Modal Network for Classification of Remote Sensing Data
http://arxiv.org/abs/2006.13806
AUTHORS:        Danfeng Hong ; Naoto Yokoya ; Gui-Song Xia ; Jocelyn Chanussot ; Xiao Xiang Zhu
HIGHLIGHT:      To this end, we propose a novel cross-modal deep-learning framework, called X-ModalNet, with three well-designed modules: self-adversarial module, interactive learning module, and label propagation module, by learning to transfer more discriminative information from a small-scale hyperspectral image (HSI) into the classification task using a large-scale MSI or SAR data.

8, TITLE:       Knee Injury Detection using MRI with Efficiently-Layered Network (ELNet)
http://arxiv.org/abs/2005.02706
AUTHORS:        Chen-Han Tsai ; Nahum Kiryati ; Eli Konen ; Iris Eshed ; Arnaldo Mayer
COMMENTS:       11 pages, 4 figures, Accepted to the Medical Imaging and Deep Learning (MIDL) Conference 2020
HIGHLIGHT:      In this work, we present the Efficiently-Layered Network (ELNet), a convolutional neural network (CNN) architecture optimized for the task of initial knee MRI diagnosis for triage.

9, TITLE:       PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning
http://arxiv.org/abs/2006.16779
AUTHORS:        Siqi Bao ; Huang He ; Fan Wang ; Hua Wu ; Haifeng Wang ; Wenquan Wu ; Zhen Guo ; Zhibin Liu ; Xinchao Xu
COMMENTS:       First four authors contributed equally to this work
HIGHLIGHT:      To build a high-quality open-domain chatbot, we introduce the effective training process of PLATO-2 via curriculum learning.

10, TITLE:      The Gambler's Problem and Beyond
http://arxiv.org/abs/2001.00102
AUTHORS:        Baoxiang Wang ; Shuai Li ; Jiajin Li ; Siu On Chan
COMMENTS:       International Conference on Learning Representations (ICLR) 2020
HIGHLIGHT:      We analyze the Gambler's problem, a simple reinforcement learning problem where the gambler has the chance to double or lose the bets until the target is reached.

11, TITLE:      Invertible generative models for inverse problems: mitigating representation error and dataset bias
http://arxiv.org/abs/1905.11672
AUTHORS:        Muhammad Asim ; Max Daniels ; Oscar Leong ; Ali Ahmed ; Paul Hand
COMMENTS:       Camera ready version for ICML 2020, paper 2655
HIGHLIGHT:      In this paper, we demonstrate that invertible neural networks, which have zero representation error by design, can be effective natural signal priors at inverse problems such as denoising, compressive sensing, and inpainting.

12, TITLE:      Program Sketching with Live Bidirectional Evaluation
http://arxiv.org/abs/1911.00583
AUTHORS:        Justin Lubin ; Nick Collins ; Cyrus Omar ; Ravi Chugh
COMMENTS:       ICFP 2020 Paper + Supplementary Appendices
HIGHLIGHT:      We present a system called Smyth for program sketching in a typed functional language whereby the concrete evaluation of ordinary assertions gives rise to input-output examples, which are then used to guide the search to complete the holes.

13, TITLE:      Broadening Label-based Argumentation Semantics with May-Must Scales (May-Must Argumentation)
http://arxiv.org/abs/2001.05730
AUTHORS:        Ryuta Arisaka ; Takayuki Ito
COMMENTS:       Changes made to the previous version. 1. Definitions of satisfaction of may/must conditions have been simplified. 2. Corrected the definition of a maximally designating labelling which is now called a maximally proper labelling instead
HIGHLIGHT:      In this work, we contemplate a way of broadening it by accommodating may- and must- conditions for an argument to be accepted or rejected, as determined by the number(s) of rejected and accepted attacking arguments.

14, TITLE:      BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models
http://arxiv.org/abs/2003.11142
AUTHORS:        Jiahui Yu ; Pengchong Jin ; Hanxiao Liu ; Gabriel Bender ; Pieter-Jan Kindermans ; Mingxing Tan ; Thomas Huang ; Xiaodan Song ; Ruoming Pang ; Quoc Le
COMMENTS:       Accepted in ECCV 2020
HIGHLIGHT:      In this work, we propose BigNAS, an approach that challenges the conventional wisdom that post-processing of the weights is necessary to get good prediction accuracies.

15, TITLE:      Angular Visual Hardness
http://arxiv.org/abs/1912.02279
AUTHORS:        Beidi Chen ; Weiyang Liu ; Zhiding Yu ; Jan Kautz ; Anshumali Shrivastava ; Animesh Garg ; Anima Anandkumar
HIGHLIGHT:      In this paper, we propose angular visual hardness (AVH), a score given by the normalized angular distance between the sample feature embedding and the target classifier to measure sample hardness.

16, TITLE:      CORD-19: The COVID-19 Open Research Dataset
http://arxiv.org/abs/2004.10706
AUTHORS:        Lucy Lu Wang ; Kyle Lo ; Yoganand Chandrasekhar ; Russell Reas ; Jiangjiang Yang ; Doug Burdick ; Darrin Eide ; Kathryn Funk ; Yannis Katsis ; Rodney Kinney ; Yunyao Li ; Ziyang Liu ; William Merrill ; Paul Mooney ; Dewey Murdick ; Devvret Rishi ; Jerry Sheehan ; Zhihong Shen ; Brandon Stilson ; Alex Wade ; Kuansan Wang ; Nancy Xin Ru Wang ; Chris Wilhelm ; Boya Xie ; Douglas Raymond ; Daniel S. Weld ; Oren Etzioni ; Sebastian Kohlmeier
COMMENTS:       ACL NLP-COVID Workshop 2020
HIGHLIGHT:      In this article, we describe the mechanics of dataset construction, highlighting challenges and key design decisions, provide an overview of how CORD-19 has been used, and describe several shared tasks built around the dataset.

17, TITLE:      Cautious Monotonicity in Case-Based Reasoning with Abstract Argumentation
http://arxiv.org/abs/2007.05284
AUTHORS:        Guilherme Paulino-Passos ; Francesca Toni
HIGHLIGHT:      In this paper, we focus on analysing the non-monotonicity properties of a regular version of $AA{\text -}CBR$ (that we call $AA{\text -}CBR_{\succeq}$).

18, TITLE:      The Fine-Grained and Parallel Complexity of Andersen's Pointer Analysis
http://arxiv.org/abs/2006.01491
AUTHORS:        Andreas Pavlogiannis ; Anders Alnor Mathiasen
HIGHLIGHT:      In this work we draw a rich fine-grained and parallel complexity landscape of APA, and present upper and lower bounds.

19, TITLE:      Exploiting Deep Generative Prior for Versatile Image Restoration and Manipulation
http://arxiv.org/abs/2003.13659
AUTHORS:        Xingang Pan ; Xiaohang Zhan ; Bo Dai ; Dahua Lin ; Chen Change Loy ; Ping Luo
COMMENTS:       Accepted to ECCV2020 as oral. 1) Precise GAN-inversion by discriminator-guided generator finetuning. 2) A versatile way for high-quality image restoration and manipulation. Code: https://github.com/XingangPan/deep-generative-prior
HIGHLIGHT:      This work presents an effective way to exploit the image prior captured by a generative adversarial network (GAN) trained on large-scale natural images.

20, TITLE:      Rapidly Adaptable Legged Robots via Evolutionary Meta-Learning
http://arxiv.org/abs/2003.01239
AUTHORS:        Xingyou Song ; Yuxiang Yang ; Krzysztof Choromanski ; Ken Caluwaerts ; Wenbo Gao ; Chelsea Finn ; Jie Tan
COMMENTS:       Published as a conference paper at International Conference on Intelligent Robots and Systems (IROS) 2020. See http://youtu.be/_QPMCDdFC3E for associated video file, and see http://github.com/google-research/google-research/tree/master/es_maml for associated code
HIGHLIGHT:      In this work, we present a new meta-learning method that allows robots to quickly adapt to changes in dynamics.

21, TITLE:      Towards High-Fidelity 3D Face Reconstruction from In-the-Wild Images Using Graph Convolutional Networks
http://arxiv.org/abs/2003.05653
AUTHORS:        Jiangke Lin ; Yi Yuan ; Tianjia Shao ; Kun Zhou
COMMENTS:       Accepted to CVPR 2020. The source code is available at https://github.com/FuxiCV/3D-Face-GCNs
HIGHLIGHT:      In this paper, we introduce a method to reconstruct 3D facial shapes with high-fidelity textures from single-view images in-the-wild, without the need to capture a large-scale face texture database.

22, TITLE:      Excessive Invariance Causes Adversarial Vulnerability
http://arxiv.org/abs/1811.00401
AUTHORS:        Jörn-Henrik Jacobsen ; Jens Behrmann ; Richard Zemel ; Matthias Bethge
HIGHLIGHT:      One core idea of adversarial example research is to reveal neural network errors under such distribution shifts.

23, TITLE:      Generalized Mean Estimation in Monte-Carlo Tree Search
http://arxiv.org/abs/1911.00384
AUTHORS:        Tuan Dam ; Pascal Klink ; Carlo D'Eramo ; Jan Peters ; Joni Pajarinen
HIGHLIGHT:      We theoretically analyze our method providing guarantees of convergence to the optimum.

24, TITLE:      Open Domain Suggestion Mining Leveraging Fine-Grained Analysis
http://arxiv.org/abs/2007.04297
AUTHORS:        Shreya Singal ; Tanishq Goel ; Shivang Chopra ; Sonika Dahiya
HIGHLIGHT:      In an attempt to overcome these challenges, we propose a two-tier pipeline that leverages Discourse Marker based oversampling and fine-grained suggestion mining techniques to retrieve suggestions from online forums.

25, TITLE:      Multi-Site Infant Brain Segmentation Algorithms: The iSeg-2019 Challenge
http://arxiv.org/abs/2007.02096
AUTHORS:        Yue Sun ; Kun Gao ; Zhengwang Wu ; Zhihao Lei ; Ying Wei ; Jun Ma ; Xiaoping Yang ; Xue Feng ; Li Zhao ; Trung Le Phan ; Jitae Shin ; Tao Zhong ; Yu Zhang ; Lequan Yu ; Caizi Li ; Ramesh Basnet ; M. Omair Ahmad ; M. N. S. Swamy ; Wenao Ma ; Qi Dou ; Toan Duc Bui ; Camilo Bermudez Noguera ; Bennett Landman ; Ian H. Gotlib ; Kathryn L. Humphreys ; Sarah Shultz ; Longchuan Li ; Sijie Niu ; Weili Lin ; Valerie Jewells ; Gang Li ; Dinggang Shen ; Li Wang
HIGHLIGHT:      To promote methodological development in the community, iSeg-2019 challenge (http://iseg2019.web.unc.edu) provides a set of 6-month infant subjects from multiple sites with different protocols/scanners for the participating methods.

26, TITLE:      LensKit for Python: Next-Generation Software for Recommender System Experiments
http://arxiv.org/abs/1809.03125
AUTHORS:        Michael D. Ekstrand
HIGHLIGHT:      In this paper, I present the next generation of the LensKit project, re-envisioning the original tool's objectives as flexible Python package for supporting recommender systems research and development.

27, TITLE:      egg: Fast and Extensible E-graphs
http://arxiv.org/abs/2004.03082
AUTHORS:        Max Willsey ; Yisu Remy Wang ; Oliver Flatt ; Chandrakana Nandi ; Pavel Panchekha ; Zachary Tatlock
COMMENTS:       25 pages, 14 figures
HIGHLIGHT:      This work contributes two techniques that make e-graphs fast and extensible, specializing them to equality saturation.

28, TITLE:      Thanks for Nothing: Predicting Zero-Valued Activations with Lightweight Convolutional Neural Networks
http://arxiv.org/abs/1909.07636
AUTHORS:        Gil Shomron ; Ron Banner ; Moran Shkolnik ; Uri Weiser
HIGHLIGHT:      Inspired by the observation that spatial correlation exists in CNN output feature maps (ofms), we propose a method to dynamically predict whether ofm activations are zero-valued or not according to their neighboring activation values, thereby avoiding zero-valued activations and reducing the number of convolution operations.

29, TITLE:      Learning-based Quantum Robust Control: Algorithm, Applications and Experiments
http://arxiv.org/abs/1702.03946
AUTHORS:        Daoyi Dong ; Xi Xing ; Hailan Ma ; Chunlin Chen ; Zhixin Liu ; Herschel Rabitz
COMMENTS:       13 pages, 10 figures and 1 table
HIGHLIGHT:      In this paper, an improved differential evolution algorithm, referred to as \emph{msMS}\_DE, is proposed to search robust fields for various quantum control problems.

30, TITLE:      Object Detection under Rainy Conditions for Autonomous Vehicles
http://arxiv.org/abs/2006.16471
AUTHORS:        Mazin Hnewa ; Hayder Radha
COMMENTS:       Accepted in IEEE Signal Processing Magazine / Special Issue on Autonomous Driving
HIGHLIGHT:      The main objective of this paper is to present a tutorial on state-of-the-art and emerging techniques that represent leading candidates for mitigating the influence of rainy conditions on an autonomous vehicle's ability to detect objects.

31, TITLE:      Semantically Plausible and Diverse 3D Human Motion Prediction
http://arxiv.org/abs/1912.08521
AUTHORS:        Sadegh Aliakbarian ; Fatemeh Sadat Saleh ; Mathieu Salzmann ; Lars Petersson ; Stephen Gould
HIGHLIGHT:      In this paper, we address both of these problems by developing a new variational framework that accounts for both diversity and semantic of the generated future motion.

32, TITLE:      SIMBA: Specific Identity Markers for Bone Age Assessment
http://arxiv.org/abs/2007.05454
AUTHORS:        Cristina González ; María Escobar ; Laura Daza ; Felipe Torres ; Gustavo Triana ; Pablo Arbeláez
COMMENTS:       Accepted at MICCAI 2020
HIGHLIGHT:      With this lack of available methods as motivation, we present SIMBA: Specific Identity Markers for Bone Age Assessment.

33, TITLE:      Learning to Learn Words from Visual Scenes
http://arxiv.org/abs/1911.11237
AUTHORS:        Dídac Surís ; Dave Epstein ; Heng Ji ; Shih-Fu Chang ; Carl Vondrick
COMMENTS:       26 pages, 12 figures
HIGHLIGHT:      We introduce a meta-learning framework that learns how to learn word representations from unconstrained scenes.

34, TITLE:      What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language models
http://arxiv.org/abs/1907.13528
AUTHORS:        Allyson Ettinger
HIGHLIGHT:      In this paper we introduce a suite of diagnostics drawn from human language experiments, which allow us to ask targeted questions about the information used by language models for generating predictions in context.

35, TITLE:      Adversarial Filters of Dataset Biases
http://arxiv.org/abs/2002.04108
AUTHORS:        Ronan Le Bras ; Swabha Swayamdipta ; Chandra Bhagavatula ; Rowan Zellers ; Matthew E. Peters ; Ashish Sabharwal ; Yejin Choi
COMMENTS:       Accepted to ICML 2020
HIGHLIGHT:      We present extensive supporting evidence that AFLite is broadly applicable for reduction of measurable dataset biases, and that models trained on the filtered datasets yield better generalization to out-of-distribution tasks.

36, TITLE:      Automated detection of corrosion in used nuclear fuel dry storage canisters using residual neural networks
http://arxiv.org/abs/2003.03241
AUTHORS:        Theodore Papamarkou ; Hayley Guy ; Bryce Kroencke ; Jordan Miller ; Preston Robinette ; Daniel Schultz ; Jacob Hinkle ; Laura Pullum ; Catherine Schuman ; Jeremy Renshaw ; Stylianos Chatzidakis
HIGHLIGHT:      This paper proposes using residual neural networks (ResNets) for real-time detection of corrosion, including iron oxide discoloration, pitting and stress corrosion cracking, in dry storage stainless steel canisters housing used nuclear fuel.

37, TITLE:      The Runtime of the Compact Genetic Algorithm on Jump Functions
http://arxiv.org/abs/1908.06527
AUTHORS:        Benjamin Doerr
COMMENTS:       Revised version of the journal version of my GECCO 2019 (arXiv:1903.10983) and FOGA 2019 (arXiv:1904.08415) papers
HIGHLIGHT:      In the first and so far only mathematical runtime analysis of an estimation-of-distribution algorithm (EDA) on a multimodal problem, Hasen\"ohrl and Sutton (GECCO 2018) showed for any $k = o(n)$ that the compact genetic algorithm (cGA) with any hypothetical population size $\mu = \Omega(ne^{4k} + n^{3.5+\varepsilon})$ with high probability finds the optimum of the $n$-dimensional jump function with jump size $k$ in time $O(\mu n^{1.5} \log n)$.

38, TITLE:      OneGAN: Simultaneous Unsupervised Learning of Conditional Image Generation, Foreground Segmentation, and Fine-Grained Clustering
http://arxiv.org/abs/1912.13471
AUTHORS:        Yaniv Benny ; Lior Wolf
COMMENTS:       To be published in the European Conference on Computer Vision (ECCV) 2020
HIGHLIGHT:      We present a method for simultaneously learning, in an unsupervised manner, (i) a conditional image generator, (ii) foreground extraction and segmentation, (iii) clustering into a two-level class hierarchy, and (iv) object removal and background completion, all done without any use of annotation.

39, TITLE:      Pick-Object-Attack: Type-Specific Adversarial Attack for Object Detection
http://arxiv.org/abs/2006.03184
AUTHORS:        Omid Mohamad Nezami ; Akshay Chaturvedi ; Mark Dras ; Utpal Garain
COMMENTS:       The paper is under consideration at Computer Vision and Image Understanding
HIGHLIGHT:      In this paper, we generate adversarial examples for object detection, which entails detecting bounding boxes around multiple objects present in the image and classifying them at the same time, making it a harder task than against image classification.

40, TITLE:      The Automation of Acceleration: AI and the Future of Society
http://arxiv.org/abs/2007.04477
AUTHORS:        Nicholas Kluge Corrêa
HIGHLIGHT:      In this article we present a review of several points, the risks and benefits of social modernization through AI, how human society has been preparing to deal with such changes, and finally, how the debate on such technologies is taking place in a way totally dominated by European and North American societies.

41, TITLE:      Rethinking RGB-D Salient Object Detection: Models, Data Sets, and Large-Scale Benchmarks
http://arxiv.org/abs/1907.06781
AUTHORS:        Deng-Ping Fan ; Zheng Lin ; Jia-Xing Zhao ; Yun Liu ; Zhao Zhang ; Qibin Hou ; Menglong Zhu ; Ming-Ming Cheng
COMMENTS:       Accepted in TNNLS20. 15 pages, 12 figures. Code: https://github.com/DengPingFan/D3NetBenchmark
HIGHLIGHT:      In this work, we fill the gap by making the following contributions to RGB-D salient object detection. (1) We carefully collect a new SIP (salient person) dataset, which consists of ~1K high-resolution images that cover diverse real-world scenes from various viewpoints, poses, occlusions, illuminations, and backgrounds.

42, TITLE:      Topic Modeling on User Stories using Word Mover's Distance
http://arxiv.org/abs/2007.05302
AUTHORS:        Kim Julian Gülle ; Nicholas Ford ; Patrick Ebel ; Florian Brokhausen ; Andreas Vogelsang
HIGHLIGHT:      In this paper, we focus on topic modeling as a means to identify topics within a large set of crowd-generated user stories and compare three approaches: (1) a traditional approach based on Latent Dirichlet Allocation, (2) a combination of word embeddings and principal component analysis, and (3) a combination of word embeddings and Word Mover's Distance.

43, TITLE:      ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions
http://arxiv.org/abs/2003.03488
AUTHORS:        Zechun Liu ; Zhiqiang Shen ; Marios Savvides ; Kwang-Ting Cheng
COMMENTS:       Accepted to ECCV 2020. Code is available at: https://github.com/liuzechun/ReActNet
HIGHLIGHT:      In this paper, we propose several ideas for enhancing a binary network to close its accuracy gap from real-valued networks without incurring any additional computational cost.

44, TITLE:      Lightweight Image Super-Resolution with Enhanced CNN
http://arxiv.org/abs/2007.04344
AUTHORS:        Chunwei Tian ; Ruibin Zhuge ; Zhihao Wu ; Yong Xu ; Wangmeng Zuo ; Chen Chen ; Chia-Wen Lin
HIGHLIGHT:      To resolve these problems, we propose a lightweight enhanced SR CNN (LESRCN-N) with three successive sub-blocks, an information extraction and enhancement block (IEEB), a reconstruction block (RB) and an information refinement block (IRB).

45, TITLE:      PoWER-BERT: Accelerating BERT Inference via Progressive Word-vector Elimination
http://arxiv.org/abs/2001.08950
AUTHORS:        Saurabh Goyal ; Anamitra R. Choudhury ; Saurabh M. Raje ; Venkatesan T. Chakaravarthy ; Yogish Sabharwal ; Ashish Verma
COMMENTS:       11 pages, 8 figures, 4 tables
HIGHLIGHT:      We develop a novel method, called PoWER-BERT, for improving the inference time of the popular BERT model, while maintaining the accuracy.

46, TITLE:      Fine-Grained Analysis of Cross-Linguistic Syntactic Divergences
http://arxiv.org/abs/2005.03436
AUTHORS:        Dmitry Nikolaev ; Ofir Arviv ; Taelin Karidi ; Neta Kenneth ; Veronika Mitnik ; Lilja Maria Saeboe ; Omri Abend
HIGHLIGHT:      We propose a framework for extracting divergence patterns for any language pair from a parallel corpus, building on Universal Dependencies. We further present a novel dataset, a manually word-aligned subset of the Parallel UD corpus in five languages, and use it to perform a detailed corpus study.

47, TITLE:      XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization
http://arxiv.org/abs/2003.11080
AUTHORS:        Junjie Hu ; Sebastian Ruder ; Aditya Siddhant ; Graham Neubig ; Orhan Firat ; Melvin Johnson
COMMENTS:       In Proceedings of the 37th International Conference on Machine Learning (ICML). July 2020
HIGHLIGHT:      To this end, we introduce the Cross-lingual TRansfer Evaluation of Multilingual Encoders XTREME benchmark, a multi-task benchmark for evaluating the cross-lingual generalization capabilities of multilingual representations across 40 languages and 9 tasks. We release the benchmark to encourage research on cross-lingual learning methods that transfer linguistic knowledge across a diverse and representative set of languages and tasks.

48, TITLE:      Difficulty Translation in Histopathology Images
http://arxiv.org/abs/2004.12535
AUTHORS:        Jerry Wei ; Arief Suriawinata ; Xiaoying Liu ; Bing Ren ; Mustafa Nasir-Moin ; Naofumi Tomita ; Jason Wei ; Saeed Hassanpour
COMMENTS:       Accepted to 2020 Artificial Intelligence in Medicine (AIME) conference. Invited for long oral presentation
HIGHLIGHT:      We propose a difficulty translation model that modifies colorectal histopathology images to be more challenging to classify.

49, TITLE:      Sentiment Analysis on Social Media Content
http://arxiv.org/abs/2007.02144
AUTHORS:        Antony Samuels ; John Mcgonical
HIGHLIGHT:      The aim of this paper is to present a model that can perform sentiment analysis of real data collected from Twitter.

50, TITLE:      Arbitrary-Oriented Object Detection with Circular Smooth Label
http://arxiv.org/abs/2003.05597
AUTHORS:        Xue Yang ; Junchi Yan
COMMENTS:       18 pages, 7 figures, 6 tables, accepted by ECCV2020
HIGHLIGHT:      In this paper, we show that existing regression-based rotation detectors suffer the problem of discontinuous boundaries, which is directly caused by angular periodicity or corner ordering.

51, TITLE:      Attention or memory? Neurointerpretable agents in space and time
http://arxiv.org/abs/2007.04862
AUTHORS:        Lennart Bramlage ; Aurelio Cortese
COMMENTS:       8 pages, 6 figures
HIGHLIGHT:      We design a model incorporating a self-attention mechanism that implements task-state representations in semantic feature-space, and test it on a battery of Atari games.

52, TITLE:      Predictive Coding Approximates Backprop along Arbitrary Computation Graphs
http://arxiv.org/abs/2006.04182
AUTHORS:        Beren Millidge ; Alexander Tschantz ; Christopher L. Buckley
COMMENTS:       Submitted to NeurIPS 2020. Updated Acknowledgements. 11/06/20: fixed typos in maths -- 11/07/20: minor corrections
HIGHLIGHT:      Here, we demonstrate that predictive coding converges asymptotically (and in practice rapidly) to exact backprop gradients on arbitrary computation graphs using only local learning rules.

53, TITLE:      A Dual Symmetric Gauss-Seidel Alternating Direction Method of Multipliers for Hyperspectral Sparse Unmixing
http://arxiv.org/abs/1902.09135
AUTHORS:        Longfei Ren ; Chengjing Wang ; Peipei Tang ; Zheng Ma
COMMENTS:       30 pages, 6 figures
HIGHLIGHT:      In this paper, we design an efficient and convergent dual symmetric Gauss-Seidel ADMM (sGS-ADMM) for hyperspectral sparse unmixing with a TV regularization term.

54, TITLE:      CPGAN: Full-Spectrum Content-Parsing Generative Adversarial Networks for Text-to-Image Synthesis
http://arxiv.org/abs/1912.08562
AUTHORS:        Jiadong Liang ; Wenjie Pei ; Feng Lu
COMMENTS:       18 pages,8 figures
HIGHLIGHT:      In this paper we circumvent this problem by focusing on parsing the content of both the input text and the synthesized image thoroughly to model the text-to-image consistency in the semantic level.

55, TITLE:      Variational Denoising Network: Toward Blind Noise Modeling and Removal
http://arxiv.org/abs/1908.11314
AUTHORS:        Zongsheng Yue ; Hongwei Yong ; Qian Zhao ; Lei Zhang ; Deyu Meng
COMMENTS:       11 pages, 4 figures
HIGHLIGHT:      In this work we propose a new variational inference method, which integrates both noise estimation and image denoising into a unique Bayesian framework, for blind image denoising.

56, TITLE:      A Convolutional Approach to Vertebrae Detection and Labelling in Whole Spine MRI
http://arxiv.org/abs/2007.02606
AUTHORS:        Rhydian Windsor ; Amir Jamaludin ; Timor Kadir ; Andrew Zisserman
COMMENTS:       Accepted full paper to Medical Image Computing and Computer Assisted Intervention 2020. 11 pages plus appendix
HIGHLIGHT:      We propose a novel convolutional method for the detection and identification of vertebrae in whole spine MRIs.

57, TITLE:      How useful is Active Learning for Image-based Plant Phenotyping?
http://arxiv.org/abs/2006.04255
AUTHORS:        Koushik Nagasubramanian ; Talukder Z. Jubery ; Fateme Fotouhi Ardakani ; Seyed Vahid Mirnezami ; Asheesh K. Singh ; Arti Singh ; Soumik Sarkar ; Baskar Ganapathysubramanian
HIGHLIGHT:      To overcome this challenge, active learning algorithms have been proposed that reduce the amount of labeling needed by deep learning models to achieve good predictive performance.

58, TITLE:      A multimodal deep learning approach for named entity recognition from social media
http://arxiv.org/abs/2001.06888
AUTHORS:        Meysam Asgari-Chenaghlu ; M. Reza Feizi-Derakhshi ; Leili Farzinvash ; M. A. Balafar ; Cina Motamed
HIGHLIGHT:      We propose two novel deep learning approaches utilizing multimodal deep learning and Transformers.

59, TITLE:      Re-thinking Co-Salient Object Detection
http://arxiv.org/abs/2007.03380
AUTHORS:        Deng-Ping Fan ; Tengpeng Li ; Zheng Lin ; Ge-Peng Ji ; Dingwen Zhang ; Ming-Ming Cheng ; Huazhu Fu ; Jianbing Shen
COMMENTS:       22pages, 18 figures. CVPR2020-CoSOD3K extension. Code: https://dpfan.net/CoSOD3K
HIGHLIGHT:      In this paper, we conduct a comprehensive study on the co-salient object detection (CoSOD) problem for images. To tackle this issue, we first introduce a new benchmark, called CoSOD3k in the wild, which requires a large amount of semantic context, making it more challenging than existing CoSOD datasets.

60, TITLE:      Edge-Aware Deep Image Deblurring
http://arxiv.org/abs/1907.02282
AUTHORS:        Zhichao Fu ; Tianlong Ma ; Yingbin Zheng ; Hao Ye ; Jing Yang ; Liang He
HIGHLIGHT:      In this paper, we resort to human visual demands of sharp edges and propose a two-phase edge-aware deep network to improve deep image deblurring.

61, TITLE:      Unsupervised Keyphrase Extraction from Scientific Publications
http://arxiv.org/abs/1808.03712
AUTHORS:        Eirini Papagiannopoulou ; Grigorios Tsoumakas
COMMENTS:       author pre-print version
HIGHLIGHT:      We propose a novel unsupervised keyphrase extraction approach that filters candidate keywords using outlier detection.

62, TITLE:      The Tree Ensemble Layer: Differentiability meets Conditional Computation
http://arxiv.org/abs/2002.07772
AUTHORS:        Hussein Hazimeh ; Natalia Ponomareva ; Petros Mol ; Zhenyu Tan ; Rahul Mazumder
COMMENTS:       ICML 2020
HIGHLIGHT:      We aim to combine these advantages by introducing a new layer for neural networks, composed of an ensemble of differentiable decision trees (a.k.a. soft trees).

63, TITLE:      EOS: a Parallel, Self-Adaptive, Multi-Population Evolutionary Algorithm for Constrained Global Optimization
http://arxiv.org/abs/2007.04681
AUTHORS:        Lorenzo Federici ; Boris Benedikter ; Alessandro Zavoli
COMMENTS:       2020 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence), Glasgow, UK
HIGHLIGHT:      This paper presents the main characteristics of the evolutionary optimization code named EOS, Evolutionary Optimization at Sapienza, and its successful application to challenging, real-world space trajectory optimization problems.

64, TITLE:      Extracting the fundamental diagram from aerial footage
http://arxiv.org/abs/2007.03227
AUTHORS:        Rafael Makrigiorgis ; Panayiotis Kolios ; Stelios Timotheou ; Theocharis Theocharides ; Christos G. Panayiotou
COMMENTS:       5 pages, 7 figures, 2020 IEEE 91st Vehicular Technology Conference (VTC2020-Spring)
HIGHLIGHT:      In this paper we devise an innovative way to obtain the fundamental diagram through aerial footage obtained from drone platforms.

65, TITLE:      Algorithmic Copywriting: Automated Generation of Health-Related Advertisements to Improve their Performance
http://arxiv.org/abs/1910.12274
AUTHORS:        Brit Youngmann ; Ran Gilad-Bachrach ; Danny Karmon ; Elad Yom-Tov
HIGHLIGHT:      Here we develop a framework, comprised of two neural networks models, that automatically generate ads.

66, TITLE:      Inferring Lower Runtime Bounds for Integer Programs
http://arxiv.org/abs/1911.01077
AUTHORS:        Florian Frohn ; Matthias Naaf ; Marc Brockschmidt ; Jürgen Giesl
HIGHLIGHT:      We present a technique to infer lower bounds on the worst-case runtime complexity of integer programs, where in contrast to earlier work, our approach is not restricted to tail-recursion.

67, TITLE:      CIRCE at SemEval-2020 Task 1: Ensembling Context-Free and Context-Dependent Word Representations
http://arxiv.org/abs/2005.06602
AUTHORS:        Martin Pömsl ; Roman Lyapin
COMMENTS:       Accepted at SemEval-2020 Task 1 @ COLING 2020. Code available at https://github.com/mpoemsl/circe
HIGHLIGHT:      We present an ensemble model that makes predictions based on context-free and context-dependent word representations.

68, TITLE:      On Saliency Maps and Adversarial Robustness
http://arxiv.org/abs/2006.07828
AUTHORS:        Puneet Mangla ; Vedant Singh ; Vineeth N Balasubramanian
COMMENTS:       Accepted at ECML-PKDD 2020, Acknowledgements added
HIGHLIGHT:      In this work, we provide a different perspective to this coupling, and provide a method, Saliency based Adversarial training (SAT), to use saliency maps to improve adversarial robustness of a model.

69, TITLE:      Automatic Detection of Coronavirus Disease (COVID-19) Using X-ray Images and Deep Convolutional Neural Networks
http://arxiv.org/abs/2003.10849
AUTHORS:        Ali Narin ; Ceren Kaya ; Ziynet Pamuk
COMMENTS:       The manuscript has 17 pages, 7 figures and 1 table
HIGHLIGHT:      In this study, three different convolutional neural network based models (ResNet50, InceptionV3 and Inception-ResNetV2) have been proposed for the detection of coronavirus pneumonia infected patient using chest X-ray radiographs.

70, TITLE:      Stereo Visual Inertial Pose Estimation Based on Feedforward-Feedback Loops
http://arxiv.org/abs/2007.02250
AUTHORS:        Shengyang Chen ; Chih-Yung Wen ; Yajing Zou ; Wu Chen
COMMENTS:       14 pages, 14 figures, 2 tables
HIGHLIGHT:      In this paper, we present a novel stereo visual inertial pose estimation method.

71, TITLE:      Attention Guided Anomaly Localization in Images
http://arxiv.org/abs/1911.08616
AUTHORS:        Shashanka Venkataramanan ; Kuan-Chuan Peng ; Rajat Vikram Singh ; Abhijit Mahalanobis
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      Without the need of anomalous training images, we propose Convolutional Adversarial Variational autoencoder with Guided Attention (CAVGA), which localizes the anomaly with a convolutional latent variable to preserve the spatial information.

72, TITLE:      A Speech Act Classifier for Persian Texts and its Application in Identifying Rumors
http://arxiv.org/abs/1901.03904
AUTHORS:        Zoleikha Jahanbakhsh-Nagadeh ; Mohammad-Reza Feizi-Derakhshi ; Arash Sharifi
COMMENTS:       Published Link: http://jscit.nit.ac.ir/article_103557.html
HIGHLIGHT:      This study presents a dictionary-based statistical technique for Persian SA recognition.

73, TITLE:      Have you forgotten? A method to assess if machine learning models have forgotten data
http://arxiv.org/abs/2004.10129
AUTHORS:        Xiao Liu ; Sotirios A Tsaftaris
COMMENTS:       Accepted by MICCAI 2020
HIGHLIGHT:      In this paper, for the first time, we want to address the challenging question of whether data have been forgotten by a model.

74, TITLE:      Towards Regulated Deep Learning
http://arxiv.org/abs/1912.13122
AUTHORS:        Andrés García-Camino
HIGHLIGHT:      The main purpose of this paper is to bring attention to Artificial Teaching (AT) and to give a tentative answer showing a proof-of-concept implementation of Regulated Deep Learning (RDL).

75, TITLE:      Population-Based Black-Box Optimization for Biological Sequence Design
http://arxiv.org/abs/2006.03227
AUTHORS:        Christof Angermueller ; David Belanger ; Andreea Gane ; Zelda Mariet ; David Dohan ; Kevin Murphy ; Lucy Colwell ; D Sculley
HIGHLIGHT:      To improve robustness, we propose Population-Based Black-Box Optimization (P3BO), which generates batches of sequences by sampling from an ensemble of methods.

76, TITLE:      DA4AD: End-to-End Deep Attention-based Visual Localization for Autonomous Driving
http://arxiv.org/abs/2003.03026
AUTHORS:        Yao Zhou ; Guowei Wan ; Shenhua Hou ; Li Yu ; Gang Wang ; Xiaofei Rui ; Shiyu Song
COMMENTS:       19 pages, 4 figures, Accepted by ECCV 2020
HIGHLIGHT:      We present a visual localization framework based on novel deep attention aware features for autonomous driving that achieves centimeter level localization accuracy.

77, TITLE:      Improved algorithm for permutation testing
http://arxiv.org/abs/2006.08473
AUTHORS:        Xiaojin Zhang
COMMENTS:       There were some mistakes for the proposal of a simpler algorithm for testing monotone pattern
HIGHLIGHT:      We study the problem of testing forbidden patterns.

78, TITLE:      Self-Supervised GAN Compression
http://arxiv.org/abs/2007.01491
AUTHORS:        Chong Yu ; Jeff Pool
COMMENTS:       The appendix for this paper is in the following repository https://gitlab.com/dxxz/Self-Supervised-GAN-Compression-Appendix
HIGHLIGHT:      In this paper, we show that a standard model compression technique, weight pruning, cannot be applied to GANs using existing methods.

79, TITLE:      Prototype Rectification for Few-Shot Learning
http://arxiv.org/abs/1911.10713
AUTHORS:        Jinlu Liu ; Liang Song ; Yongqiang Qin
COMMENTS:       ECCV 2020 Oral
HIGHLIGHT:      In this paper, we figure out two key influencing factors of the process: the intra-class bias and the cross-class bias.

80, TITLE:      A mathematical model for universal semantics
http://arxiv.org/abs/1907.12293
AUTHORS:        Weinan E ; Yajun Zhou
COMMENTS:       Main text (12 pages, 7 figures); Software Manual (ii+262 pages, 16 figures, 12 tables, available as two ancillary files). Revised according to reviewers' comments
HIGHLIGHT:      We characterize the meaning of words with language-independent numerical fingerprints, through a mathematical analysis of recurring patterns in texts.

81, TITLE:      Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks
http://arxiv.org/abs/2007.02343
AUTHORS:        Yunfei Liu ; Xingjun Ma ; James Bailey ; Feng Lu
COMMENTS:       Accepted by ECCV-2020
HIGHLIGHT:      In this paper, we present a new type of backdoor attack inspired by an important natural phenomenon: reflection.

82, TITLE:      Anatomy-Aware Siamese Network: Exploiting Semantic Asymmetry for Accurate Pelvic Fracture Detection in X-ray Images
http://arxiv.org/abs/2007.01464
AUTHORS:        Haomin Chen ; Yirui Wang ; Kang Zheng ; Weijian Li ; Chi-Tung Cheng ; Adam P. Harrison ; Jing Xiao ; Gregory D. Hager ; Le Lu ; Chien-Hung Liao ; Shun Miao
COMMENTS:       ECCV 2020 (camera-ready)
HIGHLIGHT:      In this work, we exploit semantic anatomical symmetry or asymmetry analysis in a complex CAD scenario, i.e., anterior pelvic fracture detection in trauma PXRs, where semantically pathological (refer to as fracture) and non-pathological (e.g., pose) asymmetries both occur.

83, TITLE:      Knowledge Distillation Meets Self-Supervision
http://arxiv.org/abs/2006.07114
AUTHORS:        Guodong Xu ; Ziwei Liu ; Xiaoxiao Li ; Chen Change Loy
COMMENTS:       To appear in ECCV 2020. Code is available at: https://github.com/xuguodong03/SSKD
HIGHLIGHT:      In this paper, we discuss practical ways to exploit those noisy self-supervision signals with selective transfer for distillation.

84, TITLE:      Cultural Cartography with Word Embeddings
http://arxiv.org/abs/2007.04508
AUTHORS:        Dustin S. Stoltz ; Marshall A. Taylor
HIGHLIGHT:      Word embedding models overcome this problem by constructing a standardized meaning space where words are assigned a location based on relations of similarity to, and difference from, other words based on how they are used in natural language samples.

85, TITLE:      CoSDA-ML: Multi-Lingual Code-Switching Data Augmentation for Zero-Shot Cross-Lingual NLP
http://arxiv.org/abs/2006.06402
AUTHORS:        Libo Qin ; Minheng Ni ; Yue Zhang ; Wanxiang Che
COMMENTS:       Accepted at IJCAI2020. SOLE copyright holder is IJCAI (international Joint Conferences on Artificial Intelligence), all rights reserved. http://static.ijcai.org/2020-accepted_papers.html
HIGHLIGHT:      We propose a data augmentation framework to generate multi-lingual code-switching data to fine-tune mBERT, which encourages model to align representations from source and multiple target languages once by mixing their context information.

86, TITLE:      Robust Prediction of Punctuation and Truecasing for Medical ASR
http://arxiv.org/abs/2007.02025
AUTHORS:        Monica Sunkara ; Srikanth Ronanki ; Kalpit Dixit ; Sravan Bodapati ; Katrin Kirchhoff
COMMENTS:       Accepted for ACL NLPMC workshop 2020
HIGHLIGHT:      This paper proposes a conditional joint modeling framework for prediction of punctuation and truecasing using pretrained masked language models such as BERT, BioBERT and RoBERTa.

87, TITLE:      Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision
http://arxiv.org/abs/2007.04687
AUTHORS:        Peng Wu ; Jing Liu ; Yujia Shi ; Yujia Sun ; Fangtao Shao ; Zhaoyang Wu ; Zhiwei Yang
COMMENTS:       To appear in ECCV 2020
HIGHLIGHT:      To address this problem, in this work we first release a large-scale and multi-scene dataset named XD-Violence with a total duration of 217 hours, containing 4754 untrimmed videos with audio signals and weak labels.

88, TITLE:      3D Deep Learning on Medical Images: A Review
http://arxiv.org/abs/2004.00218
AUTHORS:        Satya P. Singh ; Lipo Wang ; Sukrit Gupta ; Haveesh Goli ; Parasuraman Padmanabhan ; Balázs Gulyás
COMMENTS:       13 pages, 4 figures, 2 tables
HIGHLIGHT:      In this paper, we trace the history of how the 3D CNN was developed from its machine learning roots, give a brief mathematical description of 3D CNN and the preprocessing steps required for medical images before feeding them to 3D CNNs.

89, TITLE:      NE-LP: Normalized Entropy and Loss Prediction based Sampling for Active Learning in Chinese Word Segmentation on EHRs
http://arxiv.org/abs/1908.08419
AUTHORS:        Tingting Cai ; Zhiyuan Ma ; Hong Zheng ; Ping He ; Ju Gao ; Yangming Zhou
COMMENTS:       submitted to Neural Computing and Applications
HIGHLIGHT:      In this paper, we follow the trend and present an active learning method for CWS in EHRs.

90, TITLE:      Joints in Random Forests
http://arxiv.org/abs/2006.14937
AUTHORS:        Alvaro H. C. Correia ; Robert Peharz ; Cassio P. de Campos
HIGHLIGHT:      In this paper, we demonstrate that DTs and RFs can naturally be interpreted as generative models, by drawing a connection to Probabilistic Circuits, a prominent class of tractable probabilistic models.

91, TITLE:      Structured Landmark Detection via Topology-Adapting Deep Graph Learning
http://arxiv.org/abs/2004.08190
AUTHORS:        Weijian Li ; Yuhang Lu ; Kang Zheng ; Haofu Liao ; Chihung Lin ; Jiebo Luo ; Chi-Tung Cheng ; Jing Xiao ; Le Lu ; Chang-Fu Kuo ; Shun Miao
COMMENTS:       Accepted to ECCV-20. Camera-ready with supplementary material
HIGHLIGHT:      In this work, we present a new topology-adapting deep graph learning approach for accurate anatomical facial and medical (e.g., hand, pelvis) landmark detection.

92, TITLE:      Embedding Propagation: Smoother Manifold for Few-Shot Classification
http://arxiv.org/abs/2003.04151
AUTHORS:        Pau Rodríguez ; Issam Laradji ; Alexandre Drouin ; Alexandre Lacoste
COMMENTS:       Published at ECCV2020
HIGHLIGHT:      In this work, we propose to use embedding propagation as an unsupervised non-parametric regularizer for manifold smoothing in few-shot classification.

93, TITLE:      Input-output equations and identifiability of linear ODE models
http://arxiv.org/abs/1910.03960
AUTHORS:        Alexey Ovchinnikov ; Gleb Pogudin ; Peter Thompson
HIGHLIGHT:      In this paper, to address this, we prove identifiability of the coefficients of input-output equations for types of differential models that often appear in practice, such as linear models with one output and linear compartment models in which, from each compartment, one can reach either a leak or an input.

94, TITLE:      Towards a self-organizing pre-symbolic neural model representing sensorimotor primitives
http://arxiv.org/abs/2006.11465
AUTHORS:        Junpei Zhong ; Angelo Cangelosi ; Stefan Wermter
HIGHLIGHT:      We propose a model that relates the conceptualization of the higher-level information from visual stimuli to the development of ventral/dorsal visual streams.

95, TITLE:      No-Rainbow Problem and the Surjective Constraint Satisfaction Problem
http://arxiv.org/abs/2003.11764
AUTHORS:        Dmitriy Zhuk
HIGHLIGHT:      In this paper we show that one of the most popular variants of the SCSP, called No-Rainbow Problem, is NP-Hard.

96, TITLE:      NestFuse: An Infrared and Visible Image Fusion Architecture based on Nest Connection and Spatial/Channel Attention Models
http://arxiv.org/abs/2007.00328
AUTHORS:        Hui Li ; Xiao-Jun Wu ; Tariq Durrani
COMMENTS:       12 pages, 13 figures, 6 tables. IEEE Transactions on Instrumentation and Measurement
HIGHLIGHT:      In this paper we propose a novel method for infrared and visible image fusion where we develop nest connection-based network and spatial/channel attention models.

97, TITLE:      Neural Architecture Design for GPU-Efficient Networks
http://arxiv.org/abs/2006.14090
AUTHORS:        Ming Lin ; Hesen Chen ; Xiuyu Sun ; Qi Qian ; Hao Li ; Rong Jin
HIGHLIGHT:      To address this issue, we propose a general principle for designing GPU-efficient networks based on extensive empirical studies.

98, TITLE:      A Branching and Merging Convolutional Network with Homogeneous Filter Capsules
http://arxiv.org/abs/2001.09136
AUTHORS:        Adam Byerly ; Tatiana Kalganova ; Ian Dear
COMMENTS:       12 pages, 8 figures, 7 tables
HIGHLIGHT:      We present a convolutional neural network design with additional branches after certain convolutions so that we can extract features with differing effective receptive fields and levels of abstraction.

99, TITLE:      Segmenting Transparent Objects in the Wild
http://arxiv.org/abs/2003.13948
AUTHORS:        Enze Xie ; Wenjia Wang ; Wenhai Wang ; Mingyu Ding ; Chunhua Shen ; Ping Luo
COMMENTS:       ECCV2020 Accept
HIGHLIGHT:      To evaluate the effectiveness of Trans10K, we propose a novel boundary-aware segmentation method, termed TransLab, which exploits boundary as the clue to improve segmentation of transparent objects. To address this important problem, this work proposes a large-scale dataset for transparent object segmentation, named Trans10K, consisting of 10,428 images of real scenarios with carefully manual annotations, which are 10 times larger than the existing datasets.

100, TITLE:     Placepedia: Comprehensive Place Understanding with Multi-Faceted Annotations
http://arxiv.org/abs/2007.03777
AUTHORS:        Huaiyi Huang ; Yuqi Zhang ; Qingqiu Huang ; Zhengkui Guo ; Ziwei Liu ; Dahua Lin
COMMENTS:       To appear in ECCV 2020. Dataset is available at: https://hahehi.github.io/placepedia.html
HIGHLIGHT:      In this work, we contribute Placepedia, a large-scale place dataset with more than 35M photos from 240K unique places.

101, TITLE:     Towards a Near Universal Time Series Data Mining Tool: Introducing the Matrix Profile
http://arxiv.org/abs/1811.03064
AUTHORS:        Chin-Chia Michael Yeh
COMMENTS:       PhD dissertation (2018)
HIGHLIGHT:      In this thesis, we have introduced a near universal time series data mining tool called matrix profile which solves the all-pairs-similarity-search problem and caches the output in an easy-to-access fashion.

102, TITLE:     Self-supervised Skull Reconstruction in Brain CT Images with Decompressive Craniectomy
http://arxiv.org/abs/2007.03817
AUTHORS:        Franco Matzkin ; Virginia Newcombe ; Susan Stevenson ; Aneesh Khetani ; Tom Newman ; Richard Digby ; Andrew Stevens ; Ben Glocker ; Enzo Ferrante
COMMENTS:       Accepted for publication in MICCAI 2020. Update: Figure 1 corrected to match description
HIGHLIGHT:      We propose and compare alternative self-supervised methods where an encoder-decoder convolutional neural network (CNN) estimates the missing bone flap on post-operative CTs.

103, TITLE:     Out-of-Distribution Detection with Distance Guarantee in Deep Generative Models
http://arxiv.org/abs/2002.03328
AUTHORS:        Yufeng Zhang ; Wanwei Liu ; Zhenbang Chen ; Ji Wang ; Zhiming Liu ; Kenli Li ; Hongmei Wei
HIGHLIGHT:      Based on our theorem and the key observation, we propose an easy-to-perform method both for group and point-wise anomaly detection via estimating the total correlation of representations in DGM.

104, TITLE:     Hierarchical Deep Q-Network from Imperfect Demonstrations in Minecraft
http://arxiv.org/abs/1912.08664
AUTHORS:        Alexey Skrynnik ; Aleksey Staroverov ; Ermek Aitygulov ; Kirill Aksenov ; Vasilii Davydov ; Aleksandr I. Panov
HIGHLIGHT:      We present a structured task-dependent replay buffer and adaptive prioritizing technique that allow the HDQfD agent to gradually erase poor-quality expert data from the buffer.

105, TITLE:     Multiple Expert Brainstorming for Domain Adaptive Person Re-identification
http://arxiv.org/abs/2007.01546
AUTHORS:        Yunpeng Zhai ; Qixiang Ye ; Shijian Lu ; Mengxi Jia ; Rongrong Ji ; Yonghong Tian
COMMENTS:       Accepted by ECCV'20
HIGHLIGHT:      In this paper, we propose a multiple expert brainstorming network (MEB-Net) for domain adaptive person re-ID, opening up a promising direction about model ensemble problem under unsupervised conditions.

106, TITLE:     Open-Domain Conversational Agents: Current Progress, Open Problems, and Future Directions
http://arxiv.org/abs/2006.12442
AUTHORS:        Stephen Roller ; Y-Lan Boureau ; Jason Weston ; Antoine Bordes ; Emily Dinan ; Angela Fan ; David Gunning ; Da Ju ; Margaret Li ; Spencer Poff ; Pratik Ringshia ; Kurt Shuster ; Eric Michael Smith ; Arthur Szlam ; Jack Urbanek ; Mary Williamson
HIGHLIGHT:      We present our view of what is necessary to build an engaging open-domain conversational agent: covering the qualities of such an agent, the pieces of the puzzle that have been built so far, and the gaping holes we have not filled yet.

107, TITLE:     Validation of image-guided cochlear implant programming techniques
http://arxiv.org/abs/1909.10137
AUTHORS:        Yiyuan Zhao ; Jianing Wang ; Rui Li ; Robert F. Labadie ; Benoit M. Dawant ; Jack H. Noble
COMMENTS:       37 pages, 12 figures, 7 tables
HIGHLIGHT:      In this article, we create a ground truth dataset with conventional CT and micro-CT images of 35 temporal bone specimens to both rigorously characterize the accuracy of these two steps and assess how inaccuracies in these steps affect the overall results.

108, TITLE:     On the Approximability of Weighted Model Integration on DNF Structures
http://arxiv.org/abs/2002.06726
AUTHORS:        Ralph Abboud ; İsmail İlkan Ceylan ; Radoslav Dimitrov
COMMENTS:       To appear in Proceedings of the Seventeenth International Conference on Principles of Knowledge Representation and Reasoning (KR 2020)
HIGHLIGHT:      In this work, we study weighted model integration, a generalization of weighted model counting which involves real variables in addition to propositional variables, and pose the following question: Does weighted model integration on DNF structures admit an FPRAS?

109, TITLE:     Validation Set Evaluation can be Wrong: An Evaluator-Generator Approach for Maximizing Online Performance of Ranking in E-commerce
http://arxiv.org/abs/2003.11941
AUTHORS:        Guangda Huzhang ; Zhen-Jia Pang ; Yongqing Gao ; Yawen Liu ; Weijie Shen ; Wen-Ji Zhou ; Qing Da ; An-Xiang Zeng ; Yang Yu
HIGHLIGHT:      Validation Set Evaluation can be Wrong: An Evaluator-Generator Approach for Maximizing Online Performance of Ranking in E-commerce

110, TITLE:     Fair comparison of skin detection approaches on publicly available datasets
http://arxiv.org/abs/1802.02531
AUTHORS:        Alessandra Lumini ; Loris Nanni
HIGHLIGHT:      In this work, we investigate the most recent researches in this field and we propose a fair comparison among approaches using several different datasets.

111, TITLE:     Improving Scholarly Knowledge Representation: Evaluating BERT-based Models for Scientific Relation Classification
http://arxiv.org/abs/2004.06153
AUTHORS:        Ming Jiang ; Jennifer D'Souza ; Sören Auer ; J. Stephen Downie
HIGHLIGHT:      To this end, we present a thorough empirical evaluation on eight Bert-based classification models by focusing on two key factors: 1) Bert model variants, and 2) classification strategies.

112, TITLE:     Probabilistic Crowd GAN: Multimodal Pedestrian Trajectory Prediction using a Graph Vehicle-Pedestrian Attention Network
http://arxiv.org/abs/2006.12906
AUTHORS:        Stuart Eiffert ; Kunming Li ; Mao Shan ; Stewart Worrall ; Salah Sukkarieh ; Eduardo Nebot
COMMENTS:       Accepted for publication in IEEE Robotics and Automation Letters (RA-L) Copyright may be transferred without notice, after which this version may no longer be accessible
HIGHLIGHT:      This problem becomes increasingly complex when we consider the uncertainty and multimodality of pedestrian motion, as well as the implicit interactions between members of a crowd, including any response to a vehicle.

113, TITLE:     TLIO: Tight Learned Inertial Odometry
http://arxiv.org/abs/2007.01867
AUTHORS:        Wenxin Liu ; David Caruso ; Eddy Ilg ; Jing Dong ; Anastasios I. Mourikis ; Kostas Daniilidis ; Vijay Kumar ; Jakob Engel
COMMENTS:       Correcting graph and bibliography. Adding journal reference information and DOI, in IEEE Robotics and Automation Letters
HIGHLIGHT:      In this work we propose a tightly-coupled Extended Kalman Filter framework for IMU-only state estimation.

114, TITLE:     Bridging Knowledge Graphs to Generate Scene Graphs
http://arxiv.org/abs/2001.02314
AUTHORS:        Alireza Zareian ; Svebor Karaman ; Shih-Fu Chang
COMMENTS:       To be presented at ECCV 2020
HIGHLIGHT:      In this paper, we present a unified formulation of these two constructs, where a scene graph is seen as an image-conditioned instantiation of a commonsense knowledge graph.

115, TITLE:     Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentation
http://arxiv.org/abs/2006.16067
AUTHORS:        Jihun Yi ; Sungroh Yoon
HIGHLIGHT:      In this paper, we address the problem of image anomaly detection and segmentation.

116, TITLE:     Data-Efficient Ranking Distillation for Image Retrieval
http://arxiv.org/abs/2007.05299
AUTHORS:        Zakaria Laskar ; Juho Kannala
COMMENTS:       10 pages, 2 figures. Edited figure 7
HIGHLIGHT:      In this paper we address knowledge distillation for metric learning problems.

117, TITLE:     A Novel Attention-based Aggregation Function to Combine Vision and Language
http://arxiv.org/abs/2004.13073
AUTHORS:        Matteo Stefanini ; Marcella Cornia ; Lorenzo Baraldi ; Rita Cucchiara
COMMENTS:       ICPR 2020
HIGHLIGHT:      In this paper, we propose a novel fully-attentive reduction method for vision and language.

118, TITLE:     Domain Adaptive Relational Reasoning for 3D Multi-Organ Segmentation
http://arxiv.org/abs/2005.09120
AUTHORS:        Shuhao Fu ; Yongyi Lu ; Yan Wang ; Yuyin Zhou ; Wei Shen ; Elliot Fishman ; Alan Yuille
COMMENTS:       Accepted at MICCAL 2020
HIGHLIGHT:      In this paper, we present a novel unsupervised domain adaptation (UDA) method, named Domain Adaptive Relational Reasoning (DARR), to generalize 3D multi-organ segmentation models to medical data collected from different scanners and/or protocols (domains).

119, TITLE:     A Survey of Algorithms for Black-Box Safety Validation
http://arxiv.org/abs/2005.02979
AUTHORS:        Anthony Corso ; Robert J. Moss ; Mark Koren ; Ritchie Lee ; Mykel J. Kochenderfer
HIGHLIGHT:      We present and discuss algorithms in the domains of optimization, path planning, reinforcement learning, and importance sampling.

120, TITLE:     Hardness of Sparse Sets and Minimal Circuit Size Problem
http://arxiv.org/abs/2003.00669
AUTHORS:        Bin Fu
HIGHLIGHT:      We develop a polynomial method on finite fields to amplify the hardness of spare sets in nondeterministic time complexity classes on a randomized streaming model.

121, TITLE:     Rethinking Depthwise Separable Convolutions: How Intra-Kernel Correlations Lead to Improved MobileNets
http://arxiv.org/abs/2003.13549
AUTHORS:        Daniel Haase ; Manuel Amthor
COMMENTS:       Published at CVPR 2020. Code and models are available under https://github.com/zeiss-microscopy/BSConv
HIGHLIGHT:      We introduce blueprint separable convolutions (BSConv) as highly efficient building blocks for CNNs.

122, TITLE:     Compositional Video Synthesis with Action Graphs
http://arxiv.org/abs/2006.15327
AUTHORS:        Amir Bar ; Roei Herzig ; Xiaolong Wang ; Gal Chechik ; Trevor Darrell ; Amir Globerson
HIGHLIGHT:      Towards this end, we introduce a generative model (AG2Vid) based on Action Graphs, a natural and convenient structure that represents the dynamics of actions between objects over time.