2020.06.16.txt

==========New Papers==========
1, TITLE:       Coherent Reconstruction of Multiple Humans from a Single Image
http://arxiv.org/abs/2006.08586
AUTHORS:        Wen Jiang ; Nikos Kolotouros ; Georgios Pavlakos ; Xiaowei Zhou ; Kostas Daniilidis
COMMENTS:       CVPR 2020. Project Page: https://jiangwenpl.github.io/multiperson/
HIGHLIGHT:      In this work, we address the problem of multi-person 3D pose estimation from a single image.

2, TITLE:       Evidence-Aware Inferential Text Generation with Vector Quantised Variational AutoEncoder
http://arxiv.org/abs/2006.08101
AUTHORS:        Daya Guo ; Duyu Tang ; Nan Duan ; Jian Yin ; Daxin Jiang ; Ming Zhou
COMMENTS:       Accepted by ACL 2020
HIGHLIGHT:      To address this, we propose an approach that automatically finds evidence for an event from a large text corpus, and leverages the evidence to guide the generation of inferential texts.

3, TITLE:       Spectral DiffuserCam: lensless snapshot hyperspectral imaging with a spectral filter array
http://arxiv.org/abs/2006.08565
AUTHORS:        Kristina Monakhova ; Kyrollos Yanny ; Neerja Aggarwal ; Laura Waller
HIGHLIGHT:      In this paper, we propose a novel, compact, and inexpensive computational camera for snapshot hyperspectral imaging.

4, TITLE:       Learning Diverse and Discriminative Representations via the Principle of Maximal Coding Rate Reduction
http://arxiv.org/abs/2006.08558
AUTHORS:        Yaodong Yu ; Kwan Ho Ryan Chan ; Chong You ; Chaobing Song ; Yi Ma
HIGHLIGHT:      To learn intrinsic low-dimensional structures from high-dimensional data that most discriminate between classes, we propose the principle of Maximal Coding Rate Reduction ($\text{MCR}^2$), an information-theoretic measure that maximizes the coding rate difference between the whole dataset and the sum of each individual class.

5, TITLE:       Pipeline PSRO: A Scalable Approach for Finding Approximate Nash Equilibria in Large Games
http://arxiv.org/abs/2006.08555
AUTHORS:        Stephen McAleer ; John Lanier ; Roy Fox ; Pierre Baldi
COMMENTS:       SM and JL contributed equally
HIGHLIGHT:      We introduce Pipeline PSRO (P2SRO), the first scalable general method for finding approximate Nash equilibria in large zero-sum imperfect-information games.

6, TITLE:       Now that I can see, I can improve: Enabling data-driven finetuning of CNNs on the edge
http://arxiv.org/abs/2006.08554
AUTHORS:        Aditya Rajagopal ; Christos-Savvas Bouganis
COMMENTS:       Accepted for publication at CVPR2020 workshop - Efficient Deep Learning for Computer Vision
HIGHLIGHT:      It explores the performance gains and costs of doing so and presents an extensible open-source framework that allows the deployment of such approaches on a wide range of network architectures and devices.

7, TITLE:       Visibility Guided NMS: Efficient Boosting of Amodal Object Detection in Crowded Traffic Scenes
http://arxiv.org/abs/2006.08547
AUTHORS:        Nils Gählert ; Niklas Hanselmann ; Uwe Franke ; Joachim Denzler
COMMENTS:       Machine Learning for Autonomous Driving Workshop at the 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada
HIGHLIGHT:      Modern 2D object detection frameworks such as Yolo, SSD or Faster R-CNN predict multiple bounding boxes per object that are refined using Non-Maximum-Suppression (NMS) to suppress all but one bounding box.

8, TITLE:       Transferring Monolingual Model to Low-Resource Language: The Case of Tigrinya
http://arxiv.org/abs/2006.07698
AUTHORS:        Abrhalei Tela ; Abraham Woubie ; Ville Hautamaki
HIGHLIGHT:      In this work, we propose a cost-effective transfer learning method to adopt a strong source language model, trained from a large monolingual corpus to a low-resource language.

9, TITLE:       Sensorless Freehand 3D Ultrasound Reconstruction via Deep Contextual Learning
http://arxiv.org/abs/2006.07694
AUTHORS:        Hengtao Guo ; Sheng Xu ; Bradford Wood ; Pingkun Yan
COMMENTS:       Provisionally accepted by MICCAI 2020
HIGHLIGHT:      In this paper, we propose a deep contextual learning network (DCL-Net), which can efficiently exploit the image feature relationship between US frames and reconstruct 3D US volumes without any tracking device.

10, TITLE:      Efficient Black-Box Adversarial Attack Guided by the Distribution of Adversarial Perturbations
http://arxiv.org/abs/2006.08538
AUTHORS:        Yan Feng ; Baoyuan Wu ; Yanbo Fan ; Zhifeng Li ; Shutao Xia
HIGHLIGHT:      In this work, we propose to transform the Gaussian-distributed variable to another space through a conditional flow-based model, to enhance the capability and flexibility of capturing the intrinsic distribution of adversarial perturbations conditioned on the benign example.

11, TITLE:      Improved Conditional Flow Models for Molecule to Image Synthesis
http://arxiv.org/abs/2006.08532
AUTHORS:        Karren Yang ; Samuel Goldman ; Wengong Jin ; Alex Lu ; Regina Barzilay ; Tommi Jaakkola ; Caroline Uhler
HIGHLIGHT:      In this paper, we aim to synthesize cell microscopy images under different molecular interventions, motivated by practical applications to drug development. To evaluate our model, we propose a new set of metrics for biological image generation that are robust, interpretable, and relevant to practitioners.

12, TITLE:      Go-CaRD -- Generic, Optical Car Part Recognition and Detection: Collection, Insights, and Applications
http://arxiv.org/abs/2006.08521
AUTHORS:        Lukas Stappen ; Xinchen Du ; Vincent Karas ; Stefan Müller ; Björn W. Schuller
COMMENTS:       submitted to IEEE MMSP 2020
HIGHLIGHT:      In this paper, we present three suitable datasets as well as quantitatively and qualitatively explore the efficacy of state-of-the-art deep learning architectures for the localisation of 29 interior and exterior vehicle regions, independent of brand, model, and environment.

13, TITLE:      Pure Pattern Calculus à la de Bruijn
http://arxiv.org/abs/2006.07674
AUTHORS:        Alexis Martín ; Alejandro Ríos ; Andrés Viso
HIGHLIGHT:      This paper extends de Bruijn's ideas to properly overcome the multi-binding problem by introducing a novel presentation of PPC with bidimensional indices, in an effort to implement a prototype for a typed functional programming language based on PPC that captures path polymorphism.

14, TITLE:      The Limit of the Batch Size
http://arxiv.org/abs/2006.08517
AUTHORS:        Yang You ; Yuhui Wang ; Huan Zhang ; Zhao Zhang ; James Demmel ; Cho-Jui Hsieh
HIGHLIGHT:      In this paper, we focus on studying the limit of the batch size.

15, TITLE:      AutoGAN-Distiller: Searching to Compress Generative Adversarial Networks
http://arxiv.org/abs/2006.08198
AUTHORS:        Yonggan Fu ; Wuyang Chen ; Haotao Wang ; Haoran Li ; Yingyan Lin ; Zhangyang Wang
COMMENTS:       Accepted by ICML2020
HIGHLIGHT:      Inspired by the recent success of AutoML in deep compression, we introduce AutoML to GAN compression and develop an AutoGAN-Distiller (AGD) framework.

16, TITLE:      Extracting N-ary Cross-sentence Relations using Constrained Subsequence Kernel
http://arxiv.org/abs/2006.08185
AUTHORS:        Sachin Pawar ; Pushpak Bhattacharyya ; Girish K. Palshikar
COMMENTS:       Appeared in 20th International Conference on Computational Linguistics and Intelligent Text Processing (CICLing 2019), https://www.cicling.org/2019/
HIGHLIGHT:      We propose a new formulation of the relation extraction task where the relations are more general than intra-sentence relations in the sense that they may span multiple sentences and may have more than two arguments.

17, TITLE:      Infinite Feature Selection: A Graph-based Feature Filtering Approach
http://arxiv.org/abs/2006.08184
AUTHORS:        Giorgio Roffo ; Simone Melzi ; Umberto Castellani ; Alessandro Vinciarelli ; Marco Cristani
COMMENTS:       TPAMI PREPRINT 2020
HIGHLIGHT:      We propose a filtering feature selection framework that considers subsets of features as paths in a graph, where a node is a feature and an edge indicates pairwise (customizable) relations among features, dealing with relevance and redundancy principles.

18, TITLE:      Derivative-free global minimization for a class of multiple minima problems
http://arxiv.org/abs/2006.08181
AUTHORS:        Xiaopeng Luo ; Xin Xu ; Daoyi Dong
COMMENTS:       14 pages, 3 figures
HIGHLIGHT:      We prove that the finite-difference based derivative-free descent (FD-DFD) methods have a capability to find the global minima for a class of multiple minima problems.

19, TITLE:      Neural gradients are lognormally distributed: understanding sparse and quantized training
http://arxiv.org/abs/2006.08173
AUTHORS:        Brian Chmiel ; Liad Ben-Uri ; Moran Shkolnik ; Elad Hoffer ; Ron Banner ; Daniel Soudry
HIGHLIGHT:      Taking this into account, we suggest two methods to reduce the computational and memory burdens of neural gradients.

20, TITLE:      Learn to Effectively Explore in Context-Based Meta-RL
http://arxiv.org/abs/2006.08170
AUTHORS:        Jin Zhang ; Jianhao Wang ; Hao Hu ; Yingfeng Chen ; Changjie Fan ; Chongjie Zhang
COMMENTS:       Committed to NeurIPS 2020
HIGHLIGHT:      In this paper, we present a novel off-policy context-based meta-RL approach that efficiently learns a separate exploration policy to support fast adaptation, as well as a context-aware exploitation policy to maximize extrinsic return.

21, TITLE:      Binary DAD-Net: Binarized Driveable Area Detection Network for Autonomous Driving
http://arxiv.org/abs/2006.08178
AUTHORS:        Alexander Frickenstein ; Manoj Rohit Vemparala ; Jakob Mayr ; Naveen Shankar Nagaraja ; Christian Unger ; Federico Tombari ; Walter Stechele
COMMENTS:       IEEE International Conference on Robotics and Automation (ICRA) 2020
HIGHLIGHT:      This paper proposes a novel binarized driveable area detection network (binary DAD-Net), which uses only binary weights and activations in the encoder, the bottleneck, and the decoder part.

22, TITLE:      Dissimilarity Mixture Autoencoder for Deep Clustering
http://arxiv.org/abs/2006.08177
AUTHORS:        Juan S. Lara ; Fabio A. González
COMMENTS:       8 pages (5 additional pages for broader impact, references and supplementary material)
HIGHLIGHT:      In this paper, we introduce the Dissimilarity Mixture Autoencoder (DMAE), a novel neural network model that uses a dissimilarity function to generalize a family of density estimation and clustering methods.

23, TITLE:      Public Willingness to Get Vaccinated Against COVID-19: How AI-Developed Vaccines Can Affect Acceptance
http://arxiv.org/abs/2006.08164
AUTHORS:        Gabriel Lima ; Hyeyoung Hwang ; Chiyoung Cha ; Meeyoung Cha
HIGHLIGHT:      We used a between-subjects study design (N=572 adults in the UK and UK) to understand the public willingness towards vaccination against the novel coronavirus under various circumstances.

24, TITLE:      Filter design for small target detection on infrared imagery using normalized-cross-correlation layer
http://arxiv.org/abs/2006.08162
AUTHORS:        H. Seçkin Demir ; Erdem Akagunduz
HIGHLIGHT:      In this paper, we introduce a machine learning approach to the problem of infrared small target detection filter design.

25, TITLE:      Intelligent Decision Support System for Updating Control Plans
http://arxiv.org/abs/2006.08153
AUTHORS:        Fadwa Oukhay ; Pascale Zaraté ; Taieb Romdhane
HIGHLIGHT:      This paper proposes an intelligent DSS for quality control planning.

26, TITLE:      Group Decision Support for agriculture planning by a combination of Mathematical Model and Collaborative Tool
http://arxiv.org/abs/2006.08151
AUTHORS:        Pascale Zaraté ; Alemany Mme ; Ana Esteso Alvarez ; Amir Sakka ; Guy Camilleri
HIGHLIGHT:      In the current work the same experiment has been conducted with real business (farmers) in order to benefit from their expertise.

27, TITLE:      A VIKOR and TOPSIS focused reanalysis of the MADM methods based on logarithmic normalization
http://arxiv.org/abs/2006.08150
AUTHORS:        Sarfaraz Zolfani ; Morteza Yazdani ; Dragan Pamucar ; Pascale Zaraté
HIGHLIGHT:      This normalization method had never been applied in any MADM methods before.

28, TITLE:      Survey on Deep Multi-modal Data Analytics: Collaboration, Rivalry and Fusion
http://arxiv.org/abs/2006.08159
AUTHORS:        Yang Wang
COMMENTS:       Appearing at ACM TOMM, 26 pages
HIGHLIGHT:      In this paper, we provide a substantial overview of the existing state-of-the-arts on the filed of multi-modal data analytics from shallow to deep spaces.

29, TITLE:      Solution Subset Selection for Final Decision Making in Evolutionary Multi-Objective Optimization
http://arxiv.org/abs/2006.08156
AUTHORS:        Hisao Ishibuchi ; Lie Meng Pang ; Ke Shang
HIGHLIGHT:      In this paper, we discuss subset selection from a viewpoint of the final decision making.

30, TITLE:      Selection of an Integrated Security Area for locating a State Military Organization (SMO) based on group decision system: a multicriteria approach
http://arxiv.org/abs/2006.08155
AUTHORS:        Jean Gomes Turet ; Ana Paula Cabral Seixtas Cabral ; Pascale Zaraté
HIGHLIGHT:      Thus, this paper aims to identify the best ISA to deploy a police battalion using group decision techniques and tools.

31, TITLE:      Anomalous Motion Detection on Highway Using Deep Learning
http://arxiv.org/abs/2006.08143
AUTHORS:        Harpreet Singh ; Emily M. Hand ; Kostas Alexis
COMMENTS:       to be published in IEEE ICIP 2020
HIGHLIGHT:      We evaluate state-of-the-art deep learning anomaly detection models and propose novel variations to these methods. This paper presents a new anomaly detection dataset - the Highway Traffic Anomaly (HTA) dataset - for the problem of detecting anomalous traffic patterns from dash cam videos of vehicles on highways.

32, TITLE:      The Social Contract for AI
http://arxiv.org/abs/2006.08140
AUTHORS:        Mirka Snyder Caron ; Abhishek Gupta
COMMENTS:       Accepted paper for presentation at the IJCAI 2019 AI for Social Good workshop
HIGHLIGHT:      For the purpose of this paper, we consider that a social contract arises when there is sufficient consensus within society to adopt and implement this new technology.

33, TITLE:      Classifying degraded images over various levels of degradation
http://arxiv.org/abs/2006.08145
AUTHORS:        Kazuki Endo ; Masayuki Tanaka ; Masatoshi Okutomi
COMMENTS:       Accepted by the 27th IEEE International Conference on Image Processing (ICIP 2020)
HIGHLIGHT:      This paper proposes a convolutional neural network to classify degraded images by using a restoration network and an ensemble learning.

34, TITLE:      Emotion Recognition in Audio and Video Using Deep Neural Networks
http://arxiv.org/abs/2006.08129
AUTHORS:        Mandeep Singh ; Yuan Fang
COMMENTS:       9 pages, 9 figures, 3 tables
HIGHLIGHT:      In this work, we attempt to explore different neural networks to improve accuracy of emotion recognition.

35, TITLE:      FinBERT: A Pretrained Language Model for Financial Communications
http://arxiv.org/abs/2006.08097
AUTHORS:        Yi Yang ; Mark Christopher Siy UY ; Allen Huang
COMMENTS:       https://github.com/yya518/FinBERT
HIGHLIGHT:      In this work,we address the need by pretraining a financial domain specific BERT models, FinBERT, using a large scale of financial communication corpora.

36, TITLE:      Neural Execution Engines: Learning to Execute Subroutines
http://arxiv.org/abs/2006.08084
AUTHORS:        Yujun Yan ; Kevin Swersky ; Danai Koutra ; Parthasarathy Ranganathan ; Milad Heshemi
HIGHLIGHT:      To address the issue, we propose a learned conditional masking mechanism, which enables the model to strongly generalize far outside of its training range with near-perfect accuracy on a variety of algorithms.

37, TITLE:      Generalized Adversarially Learned Inference
http://arxiv.org/abs/2006.08089
AUTHORS:        Yatin Dandi ; Homanga Bharadhwaj ; Abhishek Kumar ; Piyush Rai
COMMENTS:       Under review
HIGHLIGHT:      Within our proposed framework, we introduce a novel set of techniques for providing self-supervised feedback to the model based on properties, such as patch-level correspondence and cycle consistency of reconstructions.

38, TITLE:      Words ranking and Hirsch index for identifying the core of the hapaxes in political texts
http://arxiv.org/abs/2006.07667
AUTHORS:        Valerio Ficcadenti ; Roy Cerqueti ; Marcel Ausloos ; Gurjeet Dhesi
HIGHLIGHT:      This paper deals with a quantitative analysis of the content of official political speeches.

39, TITLE:      Uncertainty-aware Score Distribution Learning for Action Quality Assessment
http://arxiv.org/abs/2006.07665
AUTHORS:        Yansong Tang ; Zanlin Ni ; Jiahuan Zhou ; Danyang Zhang ; Jiwen Lu ; Ying Wu ; Jie Zhou
COMMENTS:       CVPR2020
HIGHLIGHT:      To address this issue, we propose an uncertainty-aware score distribution learning (USDL) approach for action quality assessment (AQA).

40, TITLE:      Convolutional Generation of Textured 3D Meshes
http://arxiv.org/abs/2006.07660
AUTHORS:        Dario Pavllo ; Graham Spinks ; Thomas Hofmann ; Marie-Francine Moens ; Aurelien Lucchi
HIGHLIGHT:      A key contribution of our work is the encoding of the mesh and texture as 2D representations, which are semantically aligned and can be easily modeled by a 2D convolutional GAN.

41, TITLE:      APQ: Joint Search for Network Architecture, Pruning and Quantization Policy
http://arxiv.org/abs/2006.08509
AUTHORS:        Tianzhe Wang ; Kuan Wang ; Han Cai ; Ji Lin ; Zhijian Liu ; Song Han
COMMENTS:       Accepted by CVPR 2020
HIGHLIGHT:      We present APQ for efficient deep learning inference on resource-constrained hardware.

42, TITLE:      Regularized Forward-Backward Decoder for Attention Models
http://arxiv.org/abs/2006.08506
AUTHORS:        Tobias Watzel ; Ludwig Kürzinger ; Lujun Li ; Gerhard Rigoll
COMMENTS:       Under review for Interspeech 2020
HIGHLIGHT:      In this paper, we propose a novel regularization technique incorporating a second decoder during the training phase.

43, TITLE:      QD-RL: Efficient Mixing of Quality and Diversity in Reinforcement Learning
http://arxiv.org/abs/2006.08505
AUTHORS:        Geoffrey Cideron ; Thomas Pierrot ; Nicolas Perrin ; Karim Beguir ; Olivier Sigaud
HIGHLIGHT:      We propose a novel reinforcement learning algorithm,QD-RL, that incorporates the strengths of off-policy RL algorithms into Quality Diversity (QD) approaches.

44, TITLE:      Quota-based debiasing can decrease representation of already underrepresented groups
http://arxiv.org/abs/2006.07647
AUTHORS:        Ivan Smirnov ; Florian Lemmerich ; Markus Strohmaier
HIGHLIGHT:      For the case of two correlated binary attributes we show that quota-based debiasing based on a single attribute can worsen the representation of already underrepresented groups and decrease overall fairness of selection.

45, TITLE:      RoadNet-RT: High Throughput CNN Architecture and SoC Design for Real-Time Road Segmentation
http://arxiv.org/abs/2006.07644
AUTHORS:        Lin Bai ; Yecheng Lyu ; Xinming Huang
COMMENTS:       has been submitted for peer review
HIGHLIGHT:      In order to reach real-time process speed, a light-weight, high-throughput CNN architecture namely RoadNet-RT is proposed for road segmentation in this paper.

46, TITLE:      SE-MelGAN -- Speaker Agnostic Rapid Speech Enhancement
http://arxiv.org/abs/2006.07637
AUTHORS:        Luka Chkhetiani ; Levan Bejanidze
COMMENTS:       4 pages, 1 image, 1 table, 1 page for references
HIGHLIGHT:      We propose that it is possible to transfer the MelGAN's [3] robustness in learning speech features to speech enhancement and noise reduction domain without any model modification tasks.

47, TITLE:      DeepRhythm: Exposing DeepFakes with Attentional Visual Heartbeat Rhythms
http://arxiv.org/abs/2006.07634
AUTHORS:        Hua Qi ; Qing Guo ; Felix Juefei-Xu ; Xiaofei Xie ; Lei Ma ; Wei Feng ; Yang Liu ; Jianjun Zhao
COMMENTS:       11 pages, 7 figures
HIGHLIGHT:      In this work, we show that our conjecture holds true and the proposed method indeed can very effectively exposeDeepFakes by monitoring the heartbeat rhythms, which is termedasDeepRhythm.

48, TITLE:      Equivariant Neural Rendering
http://arxiv.org/abs/2006.07630
AUTHORS:        Emilien Dupont ; Miguel Angel Bautista ; Alex Colburn ; Aditya Sankar ; Carlos Guestrin ; Josh Susskind ; Qi Shan
COMMENTS:       ICML 2020 camera ready
HIGHLIGHT:      We propose a framework for learning neural scene representations directly from images, without 3D supervision. In addition, we introduce two challenging new datasets for scene representation and neural rendering, including scenes with complex lighting and backgrounds.

49, TITLE:      Special-case Algorithms for Blackbox Radical Membership, Nullstellensatz and Transcendence Degree
http://arxiv.org/abs/2006.07613
AUTHORS:        Abhibhav Garg ; Nitin Saxena
HIGHLIGHT:      We identify a useful case of these problems where practical algorithms, and improved bounds, could be given, when the transcendence degree $r$ of the input polynomials is smaller than the number of variables $n$.

50, TITLE:      DTG-Net: Differentiated Teachers Guided Self-Supervised Video Action Recognition
http://arxiv.org/abs/2006.07609
AUTHORS:        Ziming Liu ; Guangyu Gao ; A. K. Qin ; Jinyang Li
HIGHLIGHT:      To reduce such dependency, we propose a self-supervised teacher-student architecture, i.e., the Differentiated Teachers Guided self-supervised Network (DTG-Net).

51, TITLE:      HRDNet: High-resolution Detection Network for Small Objects
http://arxiv.org/abs/2006.07607
AUTHORS:        Ziming Liu ; Guangyu Gao ; Lin Sun ; Zhiyuan Fang
HIGHLIGHT:      To keep the benefits of high-resolution images without bringing up new problems, we proposed the High-Resolution Detection Network (HRDNet).

52, TITLE:      Faces à la Carte: Text-to-Face Generation via Attribute Disentanglement
http://arxiv.org/abs/2006.07606
AUTHORS:        Tianren Wang ; Teng Zhang ; Brian Lovell
COMMENTS:       8 pages, 4 figures
HIGHLIGHT:      In this paper, we propose a Text-to-Face model that not only produces images in high resolution (1024x1024) with text-to-image consistency, but also outputs multiple diverse faces to cover a wide range of unspecified facial features in a natural way.

53, TITLE:      Dynamic gesture retrieval: searching videos by human pose sequence
http://arxiv.org/abs/2006.07604
AUTHORS:        Cheng Zhang
HIGHLIGHT:      We propose a novel method for querying videos containing a designated sequence of human poses, whereas previous works only designate a single static pose.

54, TITLE:      NoPeopleAllowed: The Three-Step Approach to Weakly Supervised Semantic Segmentation
http://arxiv.org/abs/2006.07601
AUTHORS:        Mariia Dobko ; Ostap Viniavskyi ; Oles Dobosevych
COMMENTS:       This short-paper was submitted to Learning from Imperfect Data workshop at CVPR 2020
HIGHLIGHT:      We propose a novel approach to weakly supervised semantic segmentation, which consists of three consecutive steps.

55, TITLE:      Improved algorithm for permutation testing
http://arxiv.org/abs/2006.08473
AUTHORS:        Xiaojin Zhang
HIGHLIGHT:      In this paper, we provide a simple adaptive algorithm with one-sided error for testing monotone permutation.

56, TITLE:      Towards Incorporating Contextual Knowledge into the Prediction of Driving Behavior
http://arxiv.org/abs/2006.08470
AUTHORS:        Florian Wirthmüller ; Julian Schlechtriemen ; Jochen Hipp ; Manfred Reichert
COMMENTS:       the article has been accepted for publication during the 23rd IEEE Intelligent Transportation Systems Conference (ITSC), 7 pages, 6 figures, 1 table
HIGHLIGHT:      Extending former studies, we investigate how predictions are affected by external conditions.

57, TITLE:      Parametrized Fixed Points on O-Categories and Applications to Session Types
http://arxiv.org/abs/2006.08479
AUTHORS:        Ryan Kavanagh
COMMENTS:       Accepted at the 36th International Conference on Mathematical Foundations of Programming Semantics --- MFPS 2020
HIGHLIGHT:      We generalize existing techniques to define a functorial dagger operation on locally continuous functors between O-categories.

58, TITLE:      Comparing Alternative Route Planning Techniques: A Web-based Demonstration and User Study
http://arxiv.org/abs/2006.08475
AUTHORS:        Lingxiao Li ; Muhammad Aamir Cheema ; Hua Lu ; Mohammed Eunus Ali ; Adel N. Toosi
HIGHLIGHT:      Motivated by this, in this paper, we present the first user study that compares the quality of the alternative routes generated by four of the most popular existing approaches including the routes provided by Google Maps.

59, TITLE:      Algorithmically Optimal Outer Measures
http://arxiv.org/abs/2006.08468
AUTHORS:        Jack H. Lutz ; Neil Lutz
HIGHLIGHT:      We introduce global and local optimality conditions for lower semicomputable outer measures.

60, TITLE:      Oblivious and Semi-Oblivious Boundedness for Existential Rules
http://arxiv.org/abs/2006.08467
AUTHORS:        Pierre Bourhis ; Michel Leclère ; Marie-Laure Mugnier ; Sophie Tison ; Federico Ulliana ; Lily Galois
HIGHLIGHT:      We study the notion of boundedness in the context of positive existential rules, that is, whether there exists an upper bound to the depth of the chase procedure, that is independent from the initial instance.

61, TITLE:      3D-ZeF: A 3D Zebrafish Tracking Benchmark Dataset
http://arxiv.org/abs/2006.08466
AUTHORS:        Malte Pedersen ; Joakim Bruslund Haurum ; Stefan Hein Bengtson ; Thomas B. Moeslund
COMMENTS:       CVPR 2020. Project webpage: https://vap.aau.dk/3d-zef/
HIGHLIGHT:      In this work we present a novel publicly available stereo based 3D RGB dataset for multi-object zebrafish tracking, called 3D-ZeF.

62, TITLE:      Evolution of Group-Theoretic Cryptology Attacks using Hyper-heuristics
http://arxiv.org/abs/2006.08458
AUTHORS:        Matthew J. Craven ; John R. Woodward
COMMENTS:       16 pages
HIGHLIGHT:      The contribution of this paper is thus a framework to automatically build algorithms to attack cryptology problems.

63, TITLE:      Interaction Networks: Using a Reinforcement Learner to train other Machine Learning algorithms
http://arxiv.org/abs/2006.08457
AUTHORS:        Florian Dietz
HIGHLIGHT:      In this paper, thought experiments are used to explore how the additional abilities of Interaction Networks could be used to improve various existing types of neural networks.

64, TITLE:      Optimal Transport Kernels for Sequential and Parallel Neural Architecture Search
http://arxiv.org/abs/2006.07593
AUTHORS:        Vu Nguyen ; Tam Le ; Makoto Yamada ; Michael A Osborne
COMMENTS:       21 pages
HIGHLIGHT:      Building upon tree-Wasserstein (TW), which is a negative definite variant of OT, we develop a novel discrepancy for neural architectures, and demonstrate it within a Gaussian process surrogate model for the sequential NAS settings.

65, TITLE:      Attribute-aware Identity-hard Triplet Loss for Video-based Person Re-identification
http://arxiv.org/abs/2006.07597
AUTHORS:        Zhiyuan Chen ; Annan Li ; Shilu Jiang ; Yunhong Wang
HIGHLIGHT:      In this paper, we address this issue by introducing a new metric learning method called Attribute-aware Identity-hard Triplet Loss (AITL), which reduces the intra-class variation among positive samples via calculating attribute distance.

66, TITLE:      Mining Implicit Relevance Feedback from User Behavior forWeb Question Answering
http://arxiv.org/abs/2006.07581
AUTHORS:        Linjun Shou ; Shining Bo ; Feixiang Cheng ; Ming Gong ; Jian Pei ; Daxin Jiang
COMMENTS:       Accepted by KDD 2020
HIGHLIGHT:      In this paper, we make the first study to explore the correlation between user behavior and passage relevance, and propose a novel approach for mining training data for Web QA.

67, TITLE:      Adversarial Self-Supervised Contrastive Learning
http://arxiv.org/abs/2006.07589
AUTHORS:        Minseon Kim ; Jihoon Tack ; Sung Ju Hwang
COMMENTS:       Under review
HIGHLIGHT:      In this paper, we propose a novel adversarial attack for unlabeled data, which makes the model confuse the instance-level identities of the perturbed data samples.

68, TITLE:      Semantic-driven Colorization
http://arxiv.org/abs/2006.07587
AUTHORS:        Man M. Ho ; Lu Zhang ; TU Ilmenau ; Jinjia Zhou
COMMENTS:       This work is available at https://minhmanho.github.io/semantic-driven_colorization/
HIGHLIGHT:      In this study, we simulate that human-like action to firstly let our network learn to segment what is in the photo, then colorize it.

69, TITLE:      SD-RSIC: Summarization Driven Deep Remote Sensing Image Captioning
http://arxiv.org/abs/2006.08432
AUTHORS:        Gencer Sumbul ; Sonali Nayak ; Begüm Demir
HIGHLIGHT:      To overcome this limitation, in this paper we present a novel Summarization Driven Remote Sensing Image Captioning (SD-RSIC) approach.

70, TITLE:      Learning from the Scene and Borrowing from the Rich: Tackling the Long Tail in Scene Graph Generation
http://arxiv.org/abs/2006.07585
AUTHORS:        Tao He ; Lianli Gao ; Jingkuan Song ; Jianfei Cai ; Yuan-Fang Li
HIGHLIGHT:      In this paper, we tackle this issue from another two aspects: (1) scene-object interaction aiming at learning specific knowledge from a scene via an additive attention mechanism; and (2) long-tail knowledge transfer which tries to transfer the rich knowledge learned from the head into the tail.

71, TITLE:      Mitigating Face Recognition Bias via Group Adaptive Classifier
http://arxiv.org/abs/2006.07576
AUTHORS:        Sixue Gong ; Xiaoming Liu ; Anil K. Jain
HIGHLIGHT:      This work aims to learn a fair face representation, where faces of every group could be equally well-represented.

72, TITLE:      GIPFA: Generating IPA Pronunciation from Audio
http://arxiv.org/abs/2006.07573
AUTHORS:        Xavier Marjou
COMMENTS:       8 pages, 2 figures, 7 tables
HIGHLIGHT:      In this study, we instead examined the use of an Artificial Neural Network (ANN) model to automatically extract the IPA pronunciation of a word based on its audio pronunciation, hence its name Generating IPA Pronunciation From Audio (GIPFA).

73, TITLE:      Accurate Anchor Free Tracking
http://arxiv.org/abs/2006.07560
AUTHORS:        Shengyun Peng ; Yunxuan Yu ; Kun Wang ; Lei He
COMMENTS:       10 pages, 11 figures
HIGHLIGHT:      This paper develops the first Anchor Free Siamese Network (AFSN).

74, TITLE:      Unbiased Auxiliary Classifier GANs with MINE
http://arxiv.org/abs/2006.07567
AUTHORS:        Ligong Han ; Anastasis Stathopoulos ; Tao Xue ; Dimitris Metaxas
COMMENTS:       Accepted at CVPRW-20
HIGHLIGHT:      To this end, we propose an Unbiased Auxiliary GANs (UAC-GAN) that utilizes the Mutual Information Neural Estimator (MINE) to estimate the mutual information between the generated data distribution and labels.

75, TITLE:      Machine Common Sense
http://arxiv.org/abs/2006.08409
AUTHORS:        Alexander Gavrilenko ; Katerina Morozova
HIGHLIGHT:      This article deals with the aspects of modeling commonsense reasoning focusing on such domain as interpersonal interactions.

76, TITLE:      Ethical Considerations for AI Researchers
http://arxiv.org/abs/2006.07558
AUTHORS:        Kyle Dent
HIGHLIGHT:      Ethical Considerations for AI Researchers

77, TITLE:      Online Hyper-parameter Tuning in Off-policy Learning via Evolutionary Strategies
http://arxiv.org/abs/2006.07554
AUTHORS:        Yunhao Tang ; Krzysztof Choromanski
HIGHLIGHT:      In this work, we propose a framework which entails the application of Evolutionary Strategies to online hyper-parameter tuning in off-policy learning.

78, TITLE:      Sparse Separable Nonnegative Matrix Factorization
http://arxiv.org/abs/2006.07553
AUTHORS:        Nicolas Nadisic ; Arnaud Vandaele ; Jeremy E. Cohen ; Nicolas Gillis
COMMENTS:       20 pages, accepted in ECML 2020
HIGHLIGHT:      We introduce an algorithm to solve SSNMF, based on the successive nonnegative projection algorithm (SNPA, an effective algorithm for separable NMF), and an exact sparse nonnegative least squares solver.

79, TITLE:      Guided Transformer: Leveraging Multiple External Sources for Representation Learning in Conversational Search
http://arxiv.org/abs/2006.07548
AUTHORS:        Helia Hashemi ; Hamed Zamani ; W. Bruce Croft
COMMENTS:       To appear in the Proceedings of ACM SIGIR 2020. 10 pages
HIGHLIGHT:      In this paper, we enrich the representations learned by Transformer networks using a novel attention mechanism from external information sources that weights each term in the conversation.

80, TITLE:      Geo-PIFu: Geometry and Pixel Aligned Implicit Functions for Single-view Human Reconstruction
http://arxiv.org/abs/2006.08072
AUTHORS:        Tong He ; John Collomosse ; Hailin Jin ; Stefano Soatto
HIGHLIGHT:      We propose Geo-PIFu, a method to recover a 3D mesh from a monocular color image of a clothed person.

81, TITLE:      Multiple Video Frame Interpolation via Enhanced Deformable Separable Convolution
http://arxiv.org/abs/2006.08070
AUTHORS:        Xianhang Cheng ; Zhenzhong Chen
HIGHLIGHT:      In this paper, we try to solve these problems and propose a novel approach that we refer to as enhanced deformable separable convolution (EDSC) to estimate not only adaptive kernels, but also offsets, masks and biases to make the network obtain information from non-local neighborhood.

82, TITLE:      Multi-Purchase Behavior: Modeling and Optimization
http://arxiv.org/abs/2006.08055
AUTHORS:        Theja Tulabandhula ; Deeksha Sinha ; Prasoon Patidar
COMMENTS:       40 pages
HIGHLIGHT:      We present a parsimonious multi-purchase family of choice models called the BundleMVL-K family, and develop a binary search based iterative strategy that efficiently computes optimized recommendations for this model.

83, TITLE:      RasterNet: Modeling Free-Flow Speed using LiDAR and Overhead Imagery
http://arxiv.org/abs/2006.08021
AUTHORS:        Armin Hadzic ; Hunter Blanton ; Weilian Song ; Mei Chen ; Scott Workman ; Nathan Jacobs
HIGHLIGHT:      We propose a fully automated approach, RasterNet, for estimating free-flow speed without the need for explicit geometric features. To support training and evaluation, we introduce a novel dataset combining free-flow speeds of road segments, overhead imagery, and LiDAR point clouds across the state of Kentucky.

84, TITLE:      Symbolic Logic meets Machine Learning: A Brief Survey in Infinite Domains
http://arxiv.org/abs/2006.08480
AUTHORS:        Vaishak Belle
HIGHLIGHT:      In this article, we survey work that provides further evidence for the connections between logic and learning.

85, TITLE:      CompressNet: Generative Compression at Extremely Low Bitrates
http://arxiv.org/abs/2006.08003
AUTHORS:        Suraj Kiran Raman ; Aditya Ramesh ; Vijayakrishna Naganoor ; Shubham Dash ; Giridharan Kumaravelu ; Honglak Lee
HIGHLIGHT:      To address this need, we propose a novel network called CompressNet which augments a Stacked Autoencoder with a Switch Prediction Network (SAE-SPN).

86, TITLE:      UWSpeech: Speech to Speech Translation for Unwritten Languages
http://arxiv.org/abs/2006.07926
AUTHORS:        Chen Zhang ; Xu Tan ; Yi Ren ; Tao Qin ; Kejun Zhang ; Tie-Yan Liu
HIGHLIGHT:      We propose a method called XL-VAE, which enhances vector quantized variational autoencoder (VQ-VAE) with cross-lingual (XL) speech recognition, to train the converter and inverter of UWSpeech jointly.

87, TITLE:      Categorical anomaly detection in heterogeneous data using minimum description length clustering
http://arxiv.org/abs/2006.07916
AUTHORS:        James Cheney ; Xavier Gombau ; Ghita Berrada ; Sidahmed Benabderrahmane
HIGHLIGHT:      Fast and effective unsupervised anomaly detection algorithms have been proposed for categorical data based on the minimum description length (MDL) principle.

88, TITLE:      Fair Influence Maximization: A Welfare Optimization Approach
http://arxiv.org/abs/2006.07906
AUTHORS:        Aida Rahmattalabi ; Shahin Jabbari ; Himabindu Lakkaraju ; Phebe Vayanos ; Eric Rice ; Milind Tambe
HIGHLIGHT:      Algorithmic influence maximization techniques have been proposed to aid with the choice of influencers (or peer leaders) in such interventions.

89, TITLE:      Leveraging Multimodal Behavioral Analytics for Automated Job Interview Performance Assessment and Feedback
http://arxiv.org/abs/2006.07909
AUTHORS:        Anumeha Agrawal ; Rosa Anil George ; Selvan Sunitha Ravi ; Sowmya Kamath S ; Anand Kumar M
COMMENTS:       9 pages, ACL 2020
HIGHLIGHT:      We propose a multimodal analytical framework that analyzes the candidate in an interview scenario and provides feedback for predefined labels such as engagement, speaking rate, eye contact, etc.

90, TITLE:      GAN Memory with No Forgetting
http://arxiv.org/abs/2006.07543
AUTHORS:        Yulai Cong ; Miaoyun Zhao ; Jianqiao Li ; Sijia Wang ; Lawrence Carin
HIGHLIGHT:      Seeking to address the fundamental issue of memory in lifelong learning, we propose a GAN memory that is capable of realistically remembering a stream of generative processes with \emph{no} forgetting.

91, TITLE:      FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction
http://arxiv.org/abs/2006.07533
AUTHORS:        Yihao Huang ; Felix Juefei-Xu ; Run Wang ; Qing Guo ; Lei Ma ; Xiaofei Xie ; Jianwen Li ; Weikai Miao ; Yang Liu ; Geguang Pu
COMMENTS:       10 pages
HIGHLIGHT:      In this paper, we devise a simple yet powerful approach termed FakePolisher that performs shallow reconstruction of fake images through learned linear dictionary, intending to effectively and efficiently reduce the artifacts introduced during image synthesis.

92, TITLE:      Online Bayesian Goal Inference for Boundedly-Rational Planning Agents
http://arxiv.org/abs/2006.07532
AUTHORS:        Tan Zhi-Xuan ; Jordyn L. Mann ; Tom Silver ; Joshua B. Tenenbaum ; Vikash K. Mansinghka
COMMENTS:       11 pages, 5 figures (supplement: 8 pages, 8 figures)
HIGHLIGHT:      Here we present an architecture capable of inferring an agent's goals online from both optimal and non-optimal sequences of actions.

93, TITLE:      Rethinking the Value of Labels for Improving Class-Imbalanced Learning
http://arxiv.org/abs/2006.07529
AUTHORS:        Yuzhe Yang ; Zhi Xu
HIGHLIGHT:      In this work, we systematically investigate these two facets of labels.

94, TITLE:      CBR-Net: Cascade Boundary Refinement Network for Action Detection: Submission to ActivityNet Challenge 2020 (Task 1)
http://arxiv.org/abs/2006.07526
AUTHORS:        Xiang Wang ; Baiteng Ma ; Zhiwu Qing ; Yongpeng Sang ; Changxin Gao ; Shiwei Zhang ; Nong Sang
COMMENTS:       ActivityNet Challenge 2020 Temporal Action Localization (Task 1) Champion Solution (Rank 1)
HIGHLIGHT:      In this report, we present our solution for the task of temporal action localization (detection) (task 1) in ActivityNet Challenge 2020.

95, TITLE:      Self-Supervised Discovery of Anatomical Shape Landmarks
http://arxiv.org/abs/2006.07525
AUTHORS:        Riddhish Bhalodia ; Ladislav Kavan ; Ross Whitaker
COMMENTS:       Early accept at MICCAI 2020
HIGHLIGHT:      Using this assumption we propose a self-supervised, neural network approach for automatically positioning and detecting landmarks in images that can be used for subsequent analysis.

96, TITLE:      Temporal Fusion Network for Temporal Action Localization:Submission to ActivityNet Challenge 2020 (Task E)
http://arxiv.org/abs/2006.07520
AUTHORS:        Zhiwu Qing ; Xiang Wang ; Yongpeng Sang ; Changxin Gao ; Shiwei Zhang ; Nong Sang
COMMENTS:       To appear on CVPR 2020 HACS Workshop (Rank 1st)
HIGHLIGHT:      This technical report analyzes a temporal action localization method we used in the HACS competition which is hosted in Activitynet Challenge 2020.

97, TITLE:      A Multifunction Printer CUI for the Blind
http://arxiv.org/abs/2006.07519
AUTHORS:        Kyle Dent ; Kalai Ramea
HIGHLIGHT:      This paper presents a case study of our work to develop a conversational user interface for accessibility for multifunction printers (MFP).

98, TITLE:      BatVision with GCC-PHAT Features for Better Sound to Vision Predictions
http://arxiv.org/abs/2006.07995
AUTHORS:        Jesper Haahr Christensen ; Sascha Hornauer ; Stella Yu
HIGHLIGHT:      We compare and present both quantitative and qualitative improvements over our previous BatVision model.

99, TITLE:      Do Dogs have Whiskers? A New Knowledge Base of hasPart Relations
http://arxiv.org/abs/2006.07510
AUTHORS:        Sumithra Bhakthavatsalam ; Kyle Richardson ; Niket Tandon ; Peter Clark
HIGHLIGHT:      We present a new knowledge-base of hasPart relationships, extracted from a large corpus of generic statements.

100, TITLE:     Road Mapping in Low Data Environments with OpenStreetMap
http://arxiv.org/abs/2006.07993
AUTHORS:        John Kamalu ; Benjamin Choi
HIGHLIGHT:      This work investigates the viability of high resolution satellite imagery and crowd-sourced resources like OpenStreetMap in the construction of such a mapping.

101, TITLE:     Emergent Properties of Foveated Perceptual Systems
http://arxiv.org/abs/2006.07991
AUTHORS:        Arturo Deza ; Talia Konkle
COMMENTS:       A pre-print. Currently under review at the Conference on Neural Information Processing Systems (NeurIPS 2020). Themes: Foveation, Perception & Representational Learning
HIGHLIGHT:      We introduce foveated perceptual systems, inspired by human biological systems, and examine the impact that this foveation stage has on the nature and robustness of subsequently learned visual representation.

102, TITLE:     Realistic Physics Based Character Controller
http://arxiv.org/abs/2006.07508
AUTHORS:        Joe Booth ; Vladimir Ivanov
COMMENTS:       5 pages
HIGHLIGHT:      The paper aims at closing the gap between the researchers and users by introducing an open source implementation of physics based character control in Unity framework that has a low entry barrier and a steep learning curve.

103, TITLE:     GradAug: A New Regularization Method for Deep Neural Networks
http://arxiv.org/abs/2006.07989
AUTHORS:        Taojiannan Yang ; Sijie Zhu ; Chen Chen
HIGHLIGHT:      We propose a new regularization method to alleviate over-fitting in deep neural networks.

104, TITLE:     Weakly-supervised Any-shot Object Detection
http://arxiv.org/abs/2006.07502
AUTHORS:        Siddhesh Khandelwal ; Raghav Goyal ; Leonid Sigal
COMMENTS:       19 Pages, 8 Figures, 8 Tables
HIGHLIGHT:      In this work, we aim to bridge this divide by proposing an intuitive weakly-supervised model that is applicable to a range of supervision: from zero to a few instance-level samples per novel class.

105, TITLE:     Domain Generalization using Causal Matching
http://arxiv.org/abs/2006.07500
AUTHORS:        Divyat Mahajan ; Shruti Tople ; Amit Sharma
COMMENTS:       Preprint. Under review
HIGHLIGHT:      In this work, we propose a causal interpretation of domain generalization that defines domains as interventions under a data-generating process.

106, TITLE:     ShapeFlow: Learnable Deformations Among 3D Shapes
http://arxiv.org/abs/2006.07982
AUTHORS:        Chiyu "Max" Jiang ; Jingwei Huang ; Andrea Tagliasacchi ; Leonidas Guibas
HIGHLIGHT:      We present ShapeFlow, a flow-based model for learning a deformation space for entire classes of 3D shapes with large intra-class variations.

107, TITLE:     Geodesic-HOF: 3D Reconstruction Without Cutting Corners
http://arxiv.org/abs/2006.07981
AUTHORS:        Ziyun Wang ; Eric A. Mitchell ; Volkan Isler ; Daniel D. Lee
HIGHLIGHT:      To address this issue, we propose learning an image-conditioned mapping function from a canonical sampling domain to a high dimensional space where the Euclidean distance is equal to the geodesic distance on the object.

108, TITLE:     Actor-Context-Actor Relation Network for Spatio-Temporal Action Localization
http://arxiv.org/abs/2006.07976
AUTHORS:        Junting Pan ; Siyu Chen ; Zheng Shou ; Jing Shao ; Hongsheng Li
HIGHLIGHT:      In this paper, we propose to explicitly model the Actor-Context-Actor Relation, which can capture indirect high-order supportive information for effectively reasoning actors' actions in complex scenes.

109, TITLE:     Tackling Morpion Solitaire with AlphaZero-likeRanked Reward Reinforcement Learning
http://arxiv.org/abs/2006.07970
AUTHORS:        Hui Wang ; Mike Preuss ; Michael Emmerich ; Aske Plaat
COMMENTS:       4 pages, 2 figures. the first/ongoing attempt to tackle Morpion Solitaire using ranked reward reinforcement learning. submitted to SYNASC2020
HIGHLIGHT:      In this paper we take the recent impressive performance of deep self-learning reinforcement learning approaches from AlphaGo/AlphaZero as inspiration to design a searcher for Morpion Solitaire.

110, TITLE:     Relational reasoning and generalization using non-symbolic neural networks
http://arxiv.org/abs/2006.07968
AUTHORS:        Atticus Geiger ; Alexandra Carstensen Michael C. Frank ; Christopher Potts
HIGHLIGHT:      This paper revisits the question of whether equality can be learned by neural networks that do not encode explicit symbolic structure.

111, TITLE:     Meta Approach to Data Augmentation Optimization
http://arxiv.org/abs/2006.07965
AUTHORS:        Ryuichiro Hataya ; Jan Zdenek ; Kazuki Yoshizoe ; Hideki Nakayama
HIGHLIGHT:      In this paper, we propose to optimize image recognition models and data augmentation policies simultaneously to improve the performance using gradient descent.

112, TITLE:     OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfold
http://arxiv.org/abs/2006.07491
AUTHORS:        Mohamed Yousef ; Tom E. Bishop
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      We propose a novel and simple neural network module, termed \textbf{OrigamiNet}, that can augment any CTC-trained, fully convolutional single line text recognizer, to convert it into a multi-line version by providing the model with enough spatial capacity to be able to properly collapse a 2D input signal into 1D without losing information.

113, TITLE:     Understanding Unintended Memorization in Federated Learning
http://arxiv.org/abs/2006.07490
AUTHORS:        Om Thakkar ; Swaroop Ramaswamy ; Rajiv Mathews ; Françoise Beaufays
HIGHLIGHT:      In this paper, we initiate a formal study to understand the effect of different components of canonical FL on unintended memorization in trained models, comparing with the central learning setting.

114, TITLE:     A Generative Model for Joint Natural Language Understanding and Generation
http://arxiv.org/abs/2006.07499
AUTHORS:        Bo-Hsiang Tseng ; Jianpeng Cheng ; Yimai Fang ; David Vandyke
COMMENTS:       The 58th Annual Meeting of the Association for Computational Linguistics, ACL2020
HIGHLIGHT:      In this work, we propose a generative model which couples NLU and NLG through a shared latent variable.

115, TITLE:     Multi-Modal Fingerprint Presentation Attack Detection: Evaluation On A New Dataset
http://arxiv.org/abs/2006.07498
AUTHORS:        Leonidas Spinoulas ; Hengameh Mirzaalian ; Mohamed Hussein ; Wael AbdAlmageed
HIGHLIGHT:      In this work, rather than relying on legacy fingerprint images, which are widely used in the community, we study the usefulness of multiple recently introduced sensing modalities.

116, TITLE:     Open Questions in Creating Safe Open-ended AI: Tensions Between Control and Creativity
http://arxiv.org/abs/2006.07495
AUTHORS:        Adrien Ecoffet ; Jeff Clune ; Joel Lehman
HIGHLIGHT:      This paper proposes that open-ended evolution and artificial life have much to contribute towards the understanding of open-ended AI, focusing here in particular on the safety of open-ended search.

117, TITLE:     Multispectral Biometrics System Framework: Application to Presentation Attack Detection
http://arxiv.org/abs/2006.07489
AUTHORS:        Leonidas Spinoulas ; Mohamed Hussein ; David Geissbühler ; Joe Mathai ; Oswin G. Almeida ; Guillaume Clivaz ; Sébastien Marcel ; Wael AbdAlmageed
HIGHLIGHT:      In this work, we present a general framework for building a biometrics system capable of capturing multispectral data from a series of sensors synchronized with active illumination sources.

118, TITLE:     Learning-to-Learn Personalised Human Activity Recognition Models
http://arxiv.org/abs/2006.07472
AUTHORS:        Anjana Wijekoon ; Nirmalie Wiratunga
COMMENTS:       17 pages
HIGHLIGHT:      We introduce two algorithms, Personalised MAML and Personalised Relation Networks inspired by existing Meta-Learning algorithms but optimised for learning HAR models that are adaptable to any person in health and well-being applications.

119, TITLE:     ORD: Object Relationship Discovery for Visual Dialogue Generation
http://arxiv.org/abs/2006.08322
AUTHORS:        Ziwei Wang ; Zi Huang ; Yadan Luo ; Huimin Lu
HIGHLIGHT:      In this paper, we propose an object relationship discovery (ORD) framework to preserve the object interactions for visual dialogue generation.

120, TITLE:     Early Blindness Detection Based on Retinal Images Using Ensemble Learning
http://arxiv.org/abs/2006.07475
AUTHORS:        Niloy Sikder ; Md. Sanaullah Chowdhury ; Abu Shamim Mohammad Arif ; Abdullah-Al Nahid
COMMENTS:       6 pages, 22nd International Conference of Computer and Information Technology (ICCIT), 18-20 December, 2019
HIGHLIGHT:      In this study, a novel early blind detection method has been proposed based on the color information extracted from retinal images using an ensemble learning algorithm.

121, TITLE:     On the Preservation of Spatio-temporal Information in Machine Learning Applications
http://arxiv.org/abs/2006.08321
AUTHORS:        Yigit Oktar ; Mehmet Turkan
HIGHLIGHT:      In this paper, the problem of orthogonality is first investigated through conventional $k$-means of images, where images are to be processed as vectors.

122, TITLE:     Learning Causal Models Online
http://arxiv.org/abs/2006.07461
AUTHORS:        Khurram Javed ; Martha White ; Yoshua Bengio
COMMENTS:       Spurious features, causal models, online learning, random search, non-iid
HIGHLIGHT:      In this work, we take an indirect approach to discovering causal models.

123, TITLE:     TURB-Rot. A large database of 3d and 2d snapshots from turbulent rotating flows
http://arxiv.org/abs/2006.07469
AUTHORS:        L. Biferale ; F. Bonaccorso ; M. Buzzicotti ; P. Clark di Leoni
HIGHLIGHT:      We present TURB-Rot, a new open database of 3d and 2d snapshots of turbulent velocity fields, obtained by Direct Numerical Simulations (DNS) of the original Navier-Stokes equations in the presence of rotation.

124, TITLE:     Mitigating Gender Bias in Captioning Systems
http://arxiv.org/abs/2006.08315
AUTHORS:        Ruixiang Tang ; Mengnan Du ; Yuening Li ; Zirui Liu ; Xia Hu
HIGHLIGHT:      In this work, we specifically focus on the gender bias problem.

125, TITLE:     Existential Theory of the Reals Completeness of Stationary Nash Equilibria in Perfect Information Stochastic Games
http://arxiv.org/abs/2006.08314
AUTHORS:        Kristoffer Arnsfelt Hansen ; Steffan Christ Sølvsten
COMMENTS:       19 pages, 9 figures, submitted to 45th International Symposium on Mathematical Foundations of Computer Science
HIGHLIGHT:      We show that the problem of deciding whether in a multi-player perfect information recursive game (i.e. a stochastic game with terminal rewards) there exists a stationary Nash equilibrium ensuring each player a certain payoff is Existential Theory of the Reals complete.

126, TITLE:     Safety-guaranteed Reinforcement Learning based on Multi-class Support Vector Machine
http://arxiv.org/abs/2006.07446
AUTHORS:        Kwangyeon Kim ; Akshita Gupta ; Hong-Cheol Choi ; Inseok Hwang
HIGHLIGHT:      In this work, we address the problem of satisfying hard state constraints in a model-free RL setting with the deterministic system dynamics.

127, TITLE:     Algorithm for Computing Approximate Nash equilibrium in Continuous Games with Application to Continuous Blotto
http://arxiv.org/abs/2006.07443
AUTHORS:        Sam Ganzfried
HIGHLIGHT:      We present a new algorithm for computing Nash equilibrium strategies in continuous games.

128, TITLE:     Pointer Data Structure Synthesis from Answer Set Programming Specifications
http://arxiv.org/abs/2006.07440
AUTHORS:        Sarat Chandra Varanasi ; Neeraj Mittal ; Gopal Gupta
HIGHLIGHT:      We argue in this paper that ASP's reliance on negation-as-failure makes it a better formalism than those based on first-order logic for writing formal specifications.

129, TITLE:     On the Skolem Problem and Prime Powers
http://arxiv.org/abs/2006.07432
AUTHORS:        George Kenison ; Richard Lipton ; Joël Ouaknine ; James Worrell
COMMENTS:       13 pages, ISSAC 2020
HIGHLIGHT:      In this paper we consider the following specialisation of the problem: given in addition $c\in\mathbb{N}$, determine whether there exists $n\in\mathbb{N}$ of the form $n=lp^k$, with $k,l\leq c$ and $p$ any prime number, such that $u_n=0$.

130, TITLE:     Continuous Control for Searching and Planning with a Learned Model
http://arxiv.org/abs/2006.07430
AUTHORS:        Xuxi Yang ; Werner Duvaud ; Peng Wei
HIGHLIGHT:      In this paper, we provide a way and the necessary theoretical results to extend the MuZero algorithm to more generalized environments with continuous action space.

131, TITLE:     Pixel Invisibility: Detecting Objects Invisible in Color Images
http://arxiv.org/abs/2006.08383
AUTHORS:         Yongxin ;  Wang ; Duminda Wijesekera
COMMENTS:       8 pages, 7 figures, submitted to NIPS 2020
HIGHLIGHT:      We propose a novel use of cross modal knowledge distillation from color to infra-red domain using weakly-aligned image pairs from the day and construct indicators for the pixel-level invisibility based on the distances of their intermediate-level features.

132, TITLE:     DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learning
http://arxiv.org/abs/2006.08381
AUTHORS:        Kevin Ellis ; Catherine Wong ; Maxwell Nye ; Mathias Sable-Meyer ; Luc Cary ; Lucas Morales ; Luke Hewitt ; Armando Solar-Lezama ; Joshua B. Tenenbaum
HIGHLIGHT:      We present DreamCoder, a system that learns to solve problems by writing programs.

133, TITLE:     Causal Inference with Deep Causal Graphs
http://arxiv.org/abs/2006.08380
AUTHORS:        Álvaro Parafita ; Jordi Vitrià
COMMENTS:       Supplementary material can be found in https://github.com/aparafita/dcg-paper
HIGHLIGHT:      We propose Deep Causal Graphs, an abstract specification of the required functionality for a neural network to model causal distributions, and provide a model that satisfies this contract: Normalizing Causal Flows.

134, TITLE:     Catplayinginthesnow: Impact of Prior Segmentation on a Model of Visually Grounded Speech
http://arxiv.org/abs/2006.08387
AUTHORS:        William N. Havard ; Jean-Pierre Chevrot ; Laurent Besacier
HIGHLIGHT:      We introduce a simple way to introduce such information in an RNN-based model and investigate which type of boundary enables a better mapping between an image and its spoken description.

135, TITLE:     Generating Master Faces for Use in Performing Wolf Attacks on Face Recognition Systems
http://arxiv.org/abs/2006.08376
AUTHORS:        Huy H. Nguyen ; Junichi Yamagishi ; Isao Echizen ; Sébastien Marcel
COMMENTS:       Accepted to be Published in Proceedings of the 2020 International Joint Conference on Biometrics (IJCB 2020), Houston, USA
HIGHLIGHT:      In this work, we demonstrated that wolf (generic) faces, which we call "master faces," can also compromise face recognition systems and that the master face concept can be generalized in some cases.

136, TITLE:     Adaptively Meshed Video Stabilization
http://arxiv.org/abs/2006.07820
AUTHORS:        Minda Zhao ; Qiang Ling
HIGHLIGHT:      To resolve the above issues, this paper proposes an adaptively meshed method to stabilize a shaky video based on all of its feature trajectories and an adaptive blocking strategy.

137, TITLE:     Alternating ConvLSTM: Learning Force Propagation with Alternate State Updates
http://arxiv.org/abs/2006.07818
AUTHORS:        Congyue Deng ; Tai-Jiang Mu ; Shi-Min Hu
HIGHLIGHT:      In this paper, we introduce the alternating convolutional Long Short-Term Memory (Alt-ConvLSTM) that models the force propagation mechanisms in a deformable object with near-uniform material properties.

138, TITLE:     2D Image Relighting with Image-to-Image Translation
http://arxiv.org/abs/2006.07816
AUTHORS:        Paul Gafton ; Erick Maraz
COMMENTS:       12 pages, 52 Postscript figures, uses cvpr_eso.sty eso-pic.sty ruler.sty
HIGHLIGHT:      Here we provide our attempt to solve this problem using GANs.

139, TITLE:     Disentanglement for Discriminative Visual Recognition
http://arxiv.org/abs/2006.07810
AUTHORS:        Xiaofeng Liu
COMMENTS:       Manuscript for book "Recognition and perception of images" Willy
HIGHLIGHT:      In this chapter, these problems are casted as either a deep metric learning problem or an adversarial minimax game in the latent space.

140, TITLE:     Relative Pose Estimation for Stereo Rolling Shutter Cameras
http://arxiv.org/abs/2006.07807
AUTHORS:        Ke Wang ; Bin Fan ; Yuchao Dai
COMMENTS:       Accepted by International Conference on Image Processing (ICIP 2020)
HIGHLIGHT:      In this paper, we present a novel linear algorithm to estimate the 6 DoF relative pose from consecutive frames of stereo rolling shutter (RS) cameras.

141, TITLE:     Vietnamese Word Segmentation with SVM: Ambiguity Reduction and Suffix Capture
http://arxiv.org/abs/2006.07804
AUTHORS:        Duc-Vu Nguyen ; Dang Van Thin ; Kiet Van Nguyen ; Ngan Luu-Thuy Nguyen
COMMENTS:       In Proceedings of the 16th International Conference of the Pacific Association for Computational Linguistics (PACLING 2019)
HIGHLIGHT:      In this paper, we approach Vietnamese word segmentation as a binary classification by using the Support Vector Machine classifier.

142, TITLE:     Geometry-Aware Instance Segmentation with Disparity Maps
http://arxiv.org/abs/2006.07802
AUTHORS:        Cho-Ying Wu ; Xiaoyan Hu ; Michael Happold ; Qiangeng Xu ; Ulrich Neumann
COMMENTS:       CVPR 2020 Workshop of Scalability in Autonomous Driving (WSAD). Please refer to WSAD site for details
HIGHLIGHT:      Most previous works of outdoor instance segmentation for images only use color information. We collect and utilize High-Quality Driving Stereo (HQDS) dataset, using much longer baseline and focal length with higher resolution.

143, TITLE:     ReLGAN: Generalization of Consistency for GAN with Disjoint Constraints and Relative Learning of Generative Processes for Multiple Transformation Learning
http://arxiv.org/abs/2006.07809
AUTHORS:        Chiranjib Sur
HIGHLIGHT:      In this work, we have introduced a generalized scheme for consistency for GAN architectures with two new concepts of Transformation Learning (TL) and Relative Learning (ReL) for enhanced learning image transformations.

144, TITLE:     Measuring Forecasting Skill from Text
http://arxiv.org/abs/2006.07425
AUTHORS:        Shi Zong ; Alan Ritter ; Eduard Hovy
COMMENTS:       Accepted at ACL 2020
HIGHLIGHT:      In this paper we explore connections between the language people use to describe their predictions and their forecasting skill.

145, TITLE:     Defending against GAN-based Deepfake Attacks via Transformation-aware Adversarial Faces
http://arxiv.org/abs/2006.07421
AUTHORS:        Chaofei Yang ; Lei Ding ; Yiran Chen ; Hai Li
HIGHLIGHT:      This work aims to take an offensive measure to impede the generation of high-quality fake images or videos.

146, TITLE:     FinEst BERT and CroSloEngual BERT: less is more in multilingual models
http://arxiv.org/abs/2006.07890
AUTHORS:        Matej Ulčar ; Marko Robnik-Šikonja
COMMENTS:       10 pages, accepted at TSD 2020 conference
HIGHLIGHT:      We train two trilingual BERT-like models, one for Finnish, Estonian, and English, the other for Croatian, Slovenian, and English.

147, TITLE:     Team RUC_AIM3 Technical Report at Activitynet 2020 Task 2: Exploring Sequential Events Detection for Dense Video Captioning
http://arxiv.org/abs/2006.07896
AUTHORS:        Yuqing Song ; Shizhe Chen ; Yida Zhao ; Qin Jin
COMMENTS:       Winner solution in CVPR 2020 Activitynet Dense Video Captioning challenge
HIGHLIGHT:      In this work, we propose a novel and simple model for event sequence generation and explore temporal relationships of the event sequence in the video.

148, TITLE:     BI-MAML: Balanced Incremental Approach for Meta Learning
http://arxiv.org/abs/2006.07412
AUTHORS:        Yang Zheng ; Jinlin Xiang ; Kun Su ; Eli Shlizerman
COMMENTS:       Please see associated video at: https://youtu.be/4qlb-iG5SFo
HIGHLIGHT:      We present a novel Balanced Incremental Model Agnostic Meta Learning system (BI-MAML) for learning multiple tasks.

149, TITLE:     How to Avoid Being Eaten by a Grue: Structured Exploration Strategies for Textual Worlds
http://arxiv.org/abs/2006.07409
AUTHORS:        Prithviraj Ammanabrolu ; Ethan Tien ; Matthew Hausknecht ; Mark O. Riedl
HIGHLIGHT:      We introduce Q*BERT, an agent that learns to build a knowledge graph of the world by answering questions, which leads to greater sample efficiency.

150, TITLE:     Examining the Role of Mood Patterns in Predicting Self-Reported Depressive symptoms
http://arxiv.org/abs/2006.07887
AUTHORS:        Lucia Lushi Chen ; Walid Magdy ; Heather Whalley ; Maria Wolters
COMMENTS:       Accepted at The Web Science Conference 2020
HIGHLIGHT:      In this work, we attempt to enrich current technology for detecting symptoms of potential depression by constructing a 'mood profile' for social media users.

151, TITLE:     Optical Music Recognition: State of the Art and Major Challenges
http://arxiv.org/abs/2006.07885
AUTHORS:        Elona Shatri ; György Fazekas
COMMENTS:       Author manuscript for TENOR 2020 conference. 11 pages with references, 3 figures
HIGHLIGHT:      In this paper, we review relevant works in OMR, including fundamental methods and significant outcomes, and highlight different stages of the OMR pipeline.

152, TITLE:     FenceMask: A Data Augmentation Approach for Pre-extracted Image Features
http://arxiv.org/abs/2006.07877
AUTHORS:        Pu Li ; Xiangyang Li ; Xiang Long
HIGHLIGHT:      We propose a novel data augmentation method named 'FenceMask' that exhibits outstanding performance in various computer vision tasks.

153, TITLE:     Explicitly Modeled Attention Maps for Image Classification
http://arxiv.org/abs/2006.07872
AUTHORS:        Andong Tan ; Duc Tam Nguyen ; Maximilian Dax ; Matthias Niessner ; Thomas Brox
COMMENTS:       12 pages, 3 figures
HIGHLIGHT:      To mitigate this problem, we propose a novel self-attention module with explicitly modeled attention-maps using only a single learnable parameter for low computational overhead.

154, TITLE:     Comparative Evaluation of Multi-Agent Deep Reinforcement Learning Algorithms
http://arxiv.org/abs/2006.07869
AUTHORS:        Georgios Papoudakis ; Filippos Christianos ; Lukas Schäfer ; Stefano V. Albrecht
HIGHLIGHT:      In this work, we evaluate and compare three different classes of MARL algorithms (independent learners, centralised training with decentralised execution, and value decomposition) in a diverse range of multi-agent learning tasks.

155, TITLE:     Cityscapes 3D: Dataset and Benchmark for 9 DoF Vehicle Detection
http://arxiv.org/abs/2006.07864
AUTHORS:        Nils Gählert ; Nicolas Jourdan ; Marius Cordts ; Uwe Franke ; Joachim Denzler
COMMENTS:       2020 "Scalability in Autonomous Driving" CVPR Workshop
HIGHLIGHT:      To this end, we propose Cityscapes 3D, extending the original Cityscapes dataset with 3D bounding box annotations for all types of vehicles.

156, TITLE:     Continual General Chunking Problem and SyncMap
http://arxiv.org/abs/2006.07853
AUTHORS:        Danilo Vasconcellos Vargas ; Toshitake Asabuki
HIGHLIGHT:      Here, we propose a continual generalization of the chunking problem (an unsupervised problem), encompassing fixed and probabilistic chunks, discovery of temporal and causal structures and their continual variations.

157, TITLE:     Pot, kettle: Nonliteral titles aren't (natural) science
http://arxiv.org/abs/2006.07849
AUTHORS:        Mike Thelwall
COMMENTS:       Quantitative Science Studies, to appear
HIGHLIGHT:      This article investigates the prevalence of poetic expressions in journal article titles 1996-2019 in 3.3 million articles from all 27 Scopus broad fields.

158, TITLE:     An adversarial learning algorithm for mitigating gender bias in face recognition
http://arxiv.org/abs/2006.07845
AUTHORS:        Prithviraj Dhar ; Joshua Gleason ; Hossein Souri ; Carlos D. Castillo ; Rama Chellappa
HIGHLIGHT:      Therefore, we present a novel approach called `Adversarial Gender De-biasing (AGD)' to reduce the strength of gender information in face recognition features.

159, TITLE:     A Generalized Asymmetric Dual-front Model for Active Contours and Image Segmentation
http://arxiv.org/abs/2006.07839
AUTHORS:         Da~Chen ; Jack Spencer ; Jean-Marie Mirebeau ; Ke Chen ; Ming-Lei Shu ; Laurent D. Cohen
HIGHLIGHT:      In this paper, we introduce a new type of geodesic metrics that encodes the edge-based anisotropy features, the region-based homogeneity penalization and asymmetric enhancement.

160, TITLE:     Representative Committees of Peers
http://arxiv.org/abs/2006.07837
AUTHORS:        Reshef Meir ; Fedor Sandomirskiy ; Moshe Tennenholtz
HIGHLIGHT:      We show that a k-sortition (a random committee of k voters with the majority vote within the committee) leads to an outcome within the factor 1+O(1/k) of the optimal social cost for any number of voters n, any number of issues $m$, and any preference profile.

161, TITLE:     Multi-Miner: Object-Adaptive Region Mining for Weakly-Supervised Semantic Segmentation
http://arxiv.org/abs/2006.07834
AUTHORS:        Kuangqi Zhou ; Qibin Hou ; Zun Li ; Jiashi Feng
HIGHLIGHT:      In this paper, we propose a novel multi-miner framework to perform a region mining process that adapts to diverse object sizes and is thus able to mine more integral and finer object regions.

162, 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
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.

163, TITLE:     PCAAE: Principal Component Analysis Autoencoder for organising the latent space of generative networks
http://arxiv.org/abs/2006.07827
AUTHORS:        Chi-Hieu Pham ; Saïd Ladjal ; Alasdair Newson
COMMENTS:       Preprint with Appendix
HIGHLIGHT:      Drawing inspiration from principal component analysis and autoencoder, we propose the Principal Component Analysis Autoencoder (PCAAE).

164, TITLE:     Few-shot Object Detection on Remote Sensing Images
http://arxiv.org/abs/2006.07826
AUTHORS:        Jingyu Deng ; Xiang Li ; Yi Fang
COMMENTS:       12pages, 7 figures
HIGHLIGHT:      In this paper, we deal with the problem of object detection on remote sensing images.

165, TITLE:     Working with scale: 2nd place solution to Product Detection in Densely Packed Scenes [Technical Report]
http://arxiv.org/abs/2006.07825
AUTHORS:        Artem Kozlov
HIGHLIGHT:      This report describes a 2nd place solution of the detection challenge which is held within CVPR 2020 Retail-Vision workshop.

166, TITLE:     Differentiable Neural Architecture Transformation for Reproducible Architecture Improvement
http://arxiv.org/abs/2006.08231
AUTHORS:        Do-Guk Kim ; Heung-Chang Lee
HIGHLIGHT:      In this paper, we propose differentiable neural architecture transformation that is reproducible and efficient.

167, TITLE:     Piecewise-Stationary Off-Policy Optimization
http://arxiv.org/abs/2006.08236
AUTHORS:        Joey Hong ; Branislav Kveton ; Manzil Zaheer ; Yinlam Chow ; Amr Ahmed
COMMENTS:       16 pages, 2 figures
HIGHLIGHT:      To address this challenge, we study the novel problem of off-policy optimization in piecewise-stationary contextual bandits.

168, TITLE:     Finding trainable sparse networks through Neural Tangent Transfer
http://arxiv.org/abs/2006.08228
AUTHORS:        Tianlin Liu ; Friedemann Zenke
COMMENTS:       Accepted by ICML 2020
HIGHLIGHT:      In this article, we introduce Neural Tangent Transfer, a method that instead finds trainable sparse networks in a label-free manner.

169, TITLE:     Slowing Down the Weight Norm Increase in Momentum-based Optimizers
http://arxiv.org/abs/2006.08217
AUTHORS:        Byeongho Heo ; Sanghyuk Chun ; Seong Joon Oh ; Dongyoon Han ; Sangdoo Yun ; Youngjung Uh ; Jung-Woo Ha
COMMENTS:       16 pages, 9 figures. First two authors contributed equally
HIGHLIGHT:      We propose a simple and effective solution: at each iteration of momentum-based GD optimizers (e.g. SGD or Adam) applied on scale-invariant weights (e.g. Conv weights preceding a BN layer), we remove the radial component (i.e. parallel to the weight vector) from the update vector.

170, TITLE:     Structural Autoencoders Improve Representations for Generation and Transfer
http://arxiv.org/abs/2006.07796
AUTHORS:        Felix Leeb ; Yashas Annadani ; Stefan Bauer ; Bernhard Schölkopf
COMMENTS:       Submitted to NeurIPS 2020
HIGHLIGHT:      We propose a self-attention based architecture to make the encoder explicitly associate parts of the representation with parts of the input observation.

171, TITLE:     Hyper RPCA: Joint Maximum Correntropy Criterion and Laplacian Scale Mixture Modeling On-the-Fly for Moving Object Detection
http://arxiv.org/abs/2006.07795
AUTHORS:        Zerui Shao ; Yifei Pu ; Jiliu Zhou ; Bihan Wen ; Yi Zhang
HIGHLIGHT:      In this paper, we propose a novel RPCA-based model, called Hyper RPCA, to detect moving objects on the fly.

172, TITLE:     Generative 3D Part Assembly via Dynamic Graph Learning
http://arxiv.org/abs/2006.07793
AUTHORS:        Jialei Huang ; Guanqi Zhan ; Qingnan Fan ; Kaichun Mo ; Lin Shao ; Baoquan Chen ; Leonidas Guibas ; Hao Dong
HIGHLIGHT:      In this paper, we focus on the pose estimation subproblem from the vision side involving geometric and relational reasoning over the input part geometry.

173, TITLE:     PrimA6D: Rotational Primitive Reconstruction for Enhanced and Robust 6D Pose Estimation
http://arxiv.org/abs/2006.07789
AUTHORS:        MyungHwan Jeon ; Ayoung Kim
COMMENTS:       Accepted at RA-L
HIGHLIGHT:      In this paper, we introduce a rotational primitive prediction based 6D object pose estimation using a single image as an input.

174, TITLE:     Fine-grained Human Evaluation of Transformer and Recurrent Approaches to Neural Machine Translation for English-to-Chinese
http://arxiv.org/abs/2006.08297
AUTHORS:        Yuying Ye ; Antonio Toral
COMMENTS:       Accepted at the 22nd Annual Conference of the European Association for Machine Translation (EAMT 2020)
HIGHLIGHT:      This research presents a fine-grained human evaluation to compare the Transformer and recurrent approaches to neural machine translation (MT), on the translation direction English-to-Chinese.

175, TITLE:     Deep-CAPTCHA: a deep learning based CAPTCHA solver for vulnerability assessment
http://arxiv.org/abs/2006.08296
AUTHORS:        Zahra Noury ; Mahdi Rezaei
HIGHLIGHT:      In this research, we investigate a way to crack visual CAPTCHA tests by an automated deep learning based solution. In this paper, we present our customised deep neural network model, the research gaps and the existing challenges, and the solutions to overcome the issues.

176, TITLE:     Efficient Reasoning in Regular Boardgames
http://arxiv.org/abs/2006.08295
AUTHORS:        Jakub Kowalski ; Radosław Miernik ; Maksymilian Mika ; Wojciech Pawlik ; Jakub Sutowicz ; Marek Szykuła ; Andrzej Tkaczyk
COMMENTS:       IEEE Conference on Games 2020
HIGHLIGHT:      We present the technical side of reasoning in Regular Boardgames (RBG) language -- a universal General Game Playing (GGP) formalism for the class of finite deterministic games with perfect information, encoding rules in the form of regular expressions.

177, TITLE:     On the Multi-Property Extraction and Beyond
http://arxiv.org/abs/2006.08281
AUTHORS:        Tomasz Dwojak ; Michał Pietruszka ; Łukasz Borchmann ; Filip Graliński ; Jakub Chłędowski
COMMENTS:       5 pages
HIGHLIGHT:      In this paper, we investigate the Dual-source Transformer architecture on the WikiReading information extraction and machine reading comprehension dataset.

178, TITLE:     Exploration of End-to-End ASR for OpenSTT -- Russian Open Speech-to-Text Dataset
http://arxiv.org/abs/2006.08274
AUTHORS:        Andrei Andrusenko ; Aleksandr Laptev ; Ivan Medennikov
COMMENTS:       Submitted to SPECOM 2020
HIGHLIGHT:      This paper presents an exploration of end-to-end automatic speech recognition systems (ASR) for the largest open-source Russian language data set -- OpenSTT.

179, TITLE:     Randomized polynomial-time equivalence between determinant and trace-IMM equivalence tests
http://arxiv.org/abs/2006.08272
AUTHORS:        Janaky Murthy ; Vineet Nair ; Chandan Saha
COMMENTS:       36 pages, 2 figures
HIGHLIGHT:      We show that the answer is 'yes' for Det and Tr-IMM (modulo the use of randomness).

180, TITLE:     AMENet: Attentive Maps Encoder Network for Trajectory Prediction
http://arxiv.org/abs/2006.08264
AUTHORS:        Hao Cheng ; Wentong Liao ; Michael Ying Yang ; Bodo Rosenhahn ; Monika Sester
HIGHLIGHT:      To this end, we propose an end-to-end generative model named Attentive Maps Encoder Network (AMENet) for accurate and realistic multi-path trajectory prediction.

181, TITLE:     Polynomial time deterministic identity testingalgorithm for $Σ^{[3]}ΠΣΠ^{[2]}$ circuits via Edelstein-Kelly type theorem for quadratic polynomials
http://arxiv.org/abs/2006.08263
AUTHORS:        Shir Peleg ; Amir Shpilka
HIGHLIGHT:      In this work we resolve conjectures of Beecken, Mitmann and Saxena [BMS13] and Gupta [Gup14], by proving an analog of a theorem of Edelstein and Kelly for quadratic polynomials.

182, TITLE:     The PSPACE-hardness of understanding neural circuits
http://arxiv.org/abs/2006.08266
AUTHORS:        Vidya Sagar Sharma ; Piyush Srivastava
COMMENTS:       2 figures
HIGHLIGHT:      In this paper, we prove that the problems of finding minimal or minimum-size degenerate sets, and of finding the set of vital neurons, of a neural circuit given as input, are in fact PSPACE-hard.

183, TITLE:     Dermatologist vs Neural Network
http://arxiv.org/abs/2006.08254
AUTHORS:        Kaushil Mangaroliya ; Mitt Shah
HIGHLIGHT:      The model we proposed gave an accuracy of 89%.

184, TITLE:     Formalizing Falsification of Causal Structure Theories for Consciousness Across Computational Hierarchies
http://arxiv.org/abs/2006.07390
AUTHORS:        Jake R. Hanson ; Sara I. Walker
COMMENTS:       10 pages, 8 figures
HIGHLIGHT:      There is currently a global, multimillion-dollar effort to experimentally confirm or falsify neuroscience's preeminent theory of consciousness: Integrated Information Theory (IIT).

185, TITLE:     Learn to cycle: Time-consistent feature discovery for action recognition
http://arxiv.org/abs/2006.08247
AUTHORS:        Alexandros Stergiou ; Ronald Poppe
HIGHLIGHT:      In this paper, we propose a method that can learn general feature changes across time, making activations unbounded to a temporal locality, by additionally including a general notion of their learned features.

186, TITLE:     Learning Heuristic Selection with Dynamic Algorithm Configuration
http://arxiv.org/abs/2006.08246
AUTHORS:        David Speck ; André Biedenkapp ; Frank Hutter ; Robert Mattmüller ; Marius Lindauer
COMMENTS:       9 pages, 4 figures, 3 tables
HIGHLIGHT:      To learn dynamic heuristic selection, we propose an approach based on reinforcement learning and show empirically that domain-wise learned policies, which take the internal search dynamics of a planning system into account, can exceed existing approaches in terms of coverage.

187, TITLE:     Evaluating a Multi-sense Definition Generation Model for Multiple Languages
http://arxiv.org/abs/2006.07398
AUTHORS:        Arman Kabiri ; Paul Cook
COMMENTS:       To be presented orally in 23rd International Conference on Text, Speech and Dialogue (TSD 2020)
HIGHLIGHT:      In contrast, in this study, we propose a context-agnostic approach to definition modeling, based on multi-sense word embeddings, that is capable of generating multiple definitions for a target word.

188, TITLE:     The DeepFake Detection Challenge Dataset
http://arxiv.org/abs/2006.07397
AUTHORS:        Brian Dolhansky ; Joanna Bitton ; Ben Pflaum ; Jikuo Lu ; Russ Howes ; Menglin Wang ; Cristian Canton Ferrer
HIGHLIGHT:      To counter this emerging threat, we have constructed an extremely large face swap video dataset to enable the training of detection models, and organized the accompanying \textit{DeepFake Detection Challenge} (DFDC) Kaggle competition.

189, TITLE:     Cascaded deep monocular 3D human pose estimation with evolutionary training data
http://arxiv.org/abs/2006.07778
AUTHORS:        Shichao Li ; Lei Ke ; Kevin Pratama ; Yu-Wing Tai ; Chi-Keung Tang ; Kwang-Ting Cheng
HIGHLIGHT:      This paper proposes a novel data augmentation method that: (1) is scalable for synthesizing massive amount of training data (over 8 million valid 3D human poses with corresponding 2D projections) for training 2D-to-3D networks, (2) can effectively reduce dataset bias.

190, TITLE:     Domain Adaptation and Image Classification via Deep Conditional Adaptation Network
http://arxiv.org/abs/2006.07776
AUTHORS:        Pengfei Ge ; Chuan-Xian Ren ; Dao-Qing Dai ; Hong Yan
HIGHLIGHT:      To address this issue, we propose a novel unsupervised domain adaptation method, Deep Conditional Adaptation Network (DCAN), based on conditional distribution alignment of feature spaces.

191, TITLE:     Recurrent Distillation based Crowd Counting
http://arxiv.org/abs/2006.07755
AUTHORS:        Yue Gu ; Wenxi Liu
HIGHLIGHT:      Besides, leveraging our density map generation method, we propose an iterative distillation algorithm to progressively enhance our model with identical network structures, without significantly sacrificing the dimension of the output density maps.

192, TITLE:     3D Reconstruction of Novel Object Shapes from Single Images
http://arxiv.org/abs/2006.07752
AUTHORS:        Anh Thai ; Stefan Stojanov ; Vijay Upadhya ; James M. Rehg
COMMENTS:       First two authors contributed equally
HIGHLIGHT:      This paper introduces SDFNet, a novel shape prediction architecture and training approach which supports effective generalization.

193, TITLE:     Exploiting the ConvLSTM: Human Action Recognition using Raw Depth Video-Based Recurrent Neural Networks
http://arxiv.org/abs/2006.07744
AUTHORS:        Adrian Sanchez-Caballero ; David Fuentes-Jimenez ; Cristina Losada-Gutiérrez
HIGHLIGHT:      In this paper, we propose and compare, two neural networks based on the convolutional long short-term memory unit, namely ConvLSTM, with differences in the architecture and the long-term learning strategy.

194, TITLE:     3DFCNN: Real-Time Action Recognition using 3D Deep Neural Networks with Raw Depth Information
http://arxiv.org/abs/2006.07743
AUTHORS:        Adrian Sanchez-Caballero ; Sergio de López-Diz ; David Fuentes-Jimenez ; Cristina Losada-Gutiérrez ; Marta Marrón-Romera ; David Casillas-Perez ; Mohammad Ibrahim Sarker
HIGHLIGHT:      In this context, this paper describes an approach for real-time human action recognition from raw depth image-sequences, provided by an RGB-D camera.

195, TITLE:     Split-Merge Pooling
http://arxiv.org/abs/2006.07742
AUTHORS:        Omid Hosseini Jafari ; Carsten Rother
HIGHLIGHT:      In this work, we introduce Split-Merge pooling to fully preserve the spatial information without any subsampling.

196, TITLE:     DeeperGCN: All You Need to Train Deeper GCNs
http://arxiv.org/abs/2006.07739
AUTHORS:        Guohao Li ; Chenxin Xiong ; Ali Thabet ; Bernard Ghanem
COMMENTS:       This work is still working in process. More results will be updated in the future version. Project website: https://www.deepgcns.org
HIGHLIGHT:      This paper proposes DeeperGCN that is capable of successfully and reliably training very deep GCNs.

197, TITLE:     Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning
http://arxiv.org/abs/2006.07733
AUTHORS:        Jean-Bastien Grill ; Florian Strub ; Florent Altché ; Corentin Tallec ; Pierre H. Richemond ; Elena Buchatskaya ; Carl Doersch ; Bernardo Avila Pires ; Zhaohan Daniel Guo ; Mohammad Gheshlaghi Azar ; Bilal Piot ; Koray Kavukcuoglu ; Rémi Munos ; Michal Valko
HIGHLIGHT:      We introduce Bootstrap Your Own Latent (BYOL), a new approach to self-supervised image representation learning.

198, TITLE:     Through the Twitter Glass: Detecting Questions in Micro-Text
http://arxiv.org/abs/2006.07732
AUTHORS:        Kyle Dent ; Sharoda Paul
HIGHLIGHT:      This work is still preliminary, but in this paper we discuss the techniques we used and the lessons we learned.

199, TITLE:     V2E: From video frames to realistic DVS event camera streams
http://arxiv.org/abs/2006.07722
AUTHORS:        Tobi Delbruck ; Yuhuang Hu ; Zhe He
COMMENTS:       11 pages, 9 figures, technical report
HIGHLIGHT:      To help meet the increasing need for dynamic vision sensor (DVS) event camera data, we developed the v2e toolbox, which generates synthetic DVS event streams from intensity frame videos.

200, TITLE:     The Pitfalls of Simplicity Bias in Neural Networks
http://arxiv.org/abs/2006.07710
AUTHORS:        Harshay Shah ; Kaustav Tamuly ; Aditi Raghunathan ; Prateek Jain ; Praneeth Netrapalli
HIGHLIGHT:      Several works have proposed Simplicity Bias (SB)---the tendency of standard training procedures such as Stochastic Gradient Descent (SGD) to find simple models---to justify why neural networks generalize well [Arpit et al. 2017, Nakkiran et al. 2019, Valle-Perez et al. 2019].



==========Updates to Previous Papers==========
1, TITLE:       Transform and Tell: Entity-Aware News Image Captioning
http://arxiv.org/abs/2004.08070
AUTHORS:        Alasdair Tran ; Alexander Mathews ; Lexing Xie
COMMENTS:       Published in CVPR 2020. Code is available at https://github.com/alasdairtran/transform-and-tell and demo is available at https://transform-and-tell.ml
HIGHLIGHT:      We propose an end-to-end model which generates captions for images embedded in news articles. We also introduce the NYTimes800k dataset which is 70% larger than GoodNews, has higher article quality, and includes the locations of images within articles as an additional contextual cue.

2, TITLE:       FedGAN: Federated Generative Adversarial Networks for Distributed Data
http://arxiv.org/abs/2006.07228
AUTHORS:        Mohammad Rasouli ; Tao Sun ; Ram Rajagopal
COMMENTS:       23 pages, 10 figures
HIGHLIGHT:      We propose Federated Generative Adversarial Network (FedGAN) for training a GAN across distributed sources of non-independent-and-identically-distributed data sources subject to communication and privacy constraints.

3, TITLE:       Synergetic Learning Systems: Concept, Architecture, and Algorithms
http://arxiv.org/abs/2006.06367
AUTHORS:        Ping Guo ; Qian Yin
COMMENTS:       This work is based on first principle for artificial intelligence. Some terminologies and syntax errors are revised in this version
HIGHLIGHT:      Drawing on the idea that brain development is a Darwinian process of ``evolution + selection'' and the idea that the current state is a local equilibrium state of many bodies with self-organization and evolution processes driven by the temperature and gravity in our universe, in this work, we describe an artificial intelligence system called the ``Synergetic Learning Systems''.

4, TITLE:       StyleRig: Rigging StyleGAN for 3D Control over Portrait Images
http://arxiv.org/abs/2004.00121
AUTHORS:        Ayush Tewari ; Mohamed Elgharib ; Gaurav Bharaj ; Florian Bernard ; Hans-Peter Seidel ; Patrick Pérez ; Michael Zollhöfer ; Christian Theobalt
COMMENTS:       CVPR 2020 (Oral). Project page: https://gvv.mpi-inf.mpg.de/projects/StyleRig/
HIGHLIGHT:      We present the first method to provide a face rig-like control over a pretrained and fixed StyleGAN via a 3DMM.

5, TITLE:       Meta-Meta Classification for One-Shot Learning
http://arxiv.org/abs/2004.08083
AUTHORS:        Arkabandhu Chowdhury ; Dipak Chaudhari ; Swarat Chaudhuri ; Chris Jermaine
COMMENTS:       10 pages without references, 3 figures
HIGHLIGHT:      We present a new approach, called meta-meta classification, to learning in small-data settings.

6, TITLE:       Structured Knowledge Distillation for Dense Prediction
http://arxiv.org/abs/1903.04197
AUTHORS:        Yifan Liu ; Changyong Shun ; Jingdong Wang ; Chunhua Shen
COMMENTS:       v1:10 pages cvpr2019 accepted; v2:15 pages for a journal version; Code is available at: https://github.com/irfanICMLL/structure_knowledge_distillation; fix typos
HIGHLIGHT:      In this work, we consider transferring the structure information from large networks to compact ones for dense prediction tasks in computer vision.

7, TITLE:       Breast Ultrasound Computer-Aided Diagnosis Using Structure-Aware Triplet Path Networks
http://arxiv.org/abs/1908.09825
AUTHORS:        Erlei Zhang ; Zi Yang ; Stephen Seiler ; Mingli Chen ; Weiguo Lu ; Xuejun Gu
COMMENTS:       arXiv admin note: substantial text overlap with arXiv:1904.01076
HIGHLIGHT:      In this paper, a novel struc-ture-aware triplet path networks (SATPN) was designed to integrate classifica-tion and two image reconstruction tasks to achieve accurate diagnosis on US im-ages with small training dataset.

8, TITLE:       HausaMT v1.0: Towards English-Hausa Neural Machine Translation
http://arxiv.org/abs/2006.05014
AUTHORS:        Adewale Akinfaderin
COMMENTS:       Accepted at 4th Widening NLP Workshop, Annual Meeting of the Association for Computational Linguistics, ACL 2020
HIGHLIGHT:      In this paper, we curated different datasets containing Hausa-English parallel corpus for our translation.

9, TITLE:       Detection of Line Artefacts in Lung Ultrasound Images of COVID-19 Patients via Non-Convex Regularization
http://arxiv.org/abs/2005.03080
AUTHORS:        Oktay Karakuş ; Nantheera Anantrasirichai ; Amazigh Aguersif ; Stein Silva ; Adrian Basarab ; Alin Achim
COMMENTS:       16 pages, 9 figures
HIGHLIGHT:      In this paper, we present a novel method for line artefacts quantification in lung ultrasound (LUS) images of COVID-19 patients.

10, TITLE:      Attribute Restoration Framework for Anomaly Detection
http://arxiv.org/abs/1911.10676
AUTHORS:        Chaoqin Huang ; Fei Ye ; Jinkun Cao ; Maosen Li ; Ya Zhang ; Cewu Lu
HIGHLIGHT:      We here propose to break this equivalence by erasing selected attributes from the original data and reformulate it as a restoration task, where the normal and the anomalous data are expected to be distinguishable based on restoration errors.

11, TITLE:      A Neural Rendering Framework for Free-Viewpoint Relighting
http://arxiv.org/abs/1911.11530
AUTHORS:        Zhang Chen ; Anpei Chen ; Guli Zhang ; Chengyuan Wang ; Yu Ji ; Kiriakos N. Kutulakos ; Jingyi Yu
COMMENTS:       The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
HIGHLIGHT:      We present a novel Relightable Neural Renderer (RNR) for simultaneous view synthesis and relighting using multi-view image inputs.

12, TITLE:      Large-Scale Gradient-Free Deep Learning with Recursive Local Representation Alignment
http://arxiv.org/abs/2002.03911
AUTHORS:        Alexander Ororbia ; Ankur Mali ; Daniel Kifer ; C. Lee Giles
COMMENTS:       Revised submission. Additional experimental results, revisions, and substantial appendix with further details
HIGHLIGHT:      In this paper, we propose a gradient-free learning procedure, recursive local representation alignment, for training large-scale neural architectures.

13, TITLE:      On correctness of an n queens program
http://arxiv.org/abs/1909.07479
AUTHORS:        Włodzimierz Drabent
COMMENTS:       13 pages, 1 figure. This version: a simplification in the formal specification (hence in the proofs)
HIGHLIGHT:      Another purpose of the paper is to present an example of precise declarative reasoning about the semantics of a logic program.

14, TITLE:      Lower Bounds for Non-Elitist Evolutionary Algorithms via Negative Multiplicative Drift
http://arxiv.org/abs/2005.00853
AUTHORS:        Benjamin Doerr
COMMENTS:       Full version of a paper accepted for publication in the proceedings of PPSN 2020. Difference to the previous version: Added a section on standard bit mutation with random mutation rate. Additional difference to the conference version: All proofs are given since ArXiV does not have a page limit
HIGHLIGHT:      We propose a simple negative drift theorem for multiplicative drift scenarios and show that it can simplify existing analyses.

15, TITLE:      Information Extraction of Clinical Trial Eligibility Criteria
http://arxiv.org/abs/2006.07296
AUTHORS:        Yitong Tseo ; M. I. Salkola ; Ahmed Mohamed ; Anuj Kumar ; Freddy Abnousi
COMMENTS:       4 pages
HIGHLIGHT:      In this paper, we investigate an information extraction (IE) approach for grounding criteria from trials in ClinicalTrials.gov to a shared knowledge base.

16, TITLE:      Towards Hierarchical Importance Attribution: Explaining Compositional Semantics for Neural Sequence Models
http://arxiv.org/abs/1911.06194
AUTHORS:        Xisen Jin ; Zhongyu Wei ; Junyi Du ; Xiangyang Xue ; Xiang Ren
COMMENTS:       ICLR 2020
HIGHLIGHT:      In this paper, we start by proposing a formal and general way to quantify the importance of each word and phrase.

17, TITLE:      MadNet: Using a MAD Optimization for Defending Against Adversarial Attacks
http://arxiv.org/abs/1911.00870
AUTHORS:        Shai Rozenberg ; Gal Elidan ; Ran El-Yaniv
HIGHLIGHT:      Inspired by the certificate defense approach, we propose a maximal adversarial distortion (MAD) optimization method for robustifying deep networks.

18, TITLE:      Existence and Complexity of Approximate Equilibria in Weighted Congestion Games
http://arxiv.org/abs/2002.07466
AUTHORS:        George Christodoulou ; Martin Gairing ; Yiannis Giannakopoulos ; Diogo Poças ; Clara Waldmann
HIGHLIGHT:      We study the existence of approximate pure Nash equilibria ($\alpha$-PNE) in weighted atomic congestion games with polynomial cost functions of maximum degree $d$.

19, TITLE:      Star-convex Polyhedra for 3D Object Detection and Segmentation in Microscopy
http://arxiv.org/abs/1908.03636
AUTHORS:        Martin Weigert ; Uwe Schmidt ; Robert Haase ; Ko Sugawara ; Gene Myers
COMMENTS:       Conference paper at WACV 2020
HIGHLIGHT:      In this paper, we adopt and extend this approach to 3D volumes by using star-convex polyhedra to represent cell nuclei and similar shapes.

20, TITLE:      Selectivity considered harmful: evaluating the causal impact of class selectivity in DNNs
http://arxiv.org/abs/2003.01262
AUTHORS:        Matthew L. Leavitt ; Ari Morcos
HIGHLIGHT:      We investigated the causal impact of class selectivity on network function by directly regularizing for or against class selectivity.

21, TITLE:      Learning Credal Sum-Product Networks
http://arxiv.org/abs/1901.05847
AUTHORS:        Amelie Levray ; Vaishak Belle
COMMENTS:       Accepted to AKBC 2020
HIGHLIGHT:      In this work, we are interested in how such representations can be learnt and thus study how the computational machinery underlying tractable learning and inference can be generalized for imprecise probabilities.

22, TITLE:      Parametrised Complexity of Model Checking and Satisfiability in Propositional Dependence Logic
http://arxiv.org/abs/1904.06107
AUTHORS:        Yasir Mahmood ; Arne Meier
COMMENTS:       Update includes refined results
HIGHLIGHT:      In this paper, we initiate a systematic study of the parametrised complexity in the field of Dependence Logics which finds its origin in the Dependence Logic of V\"a\"an\"anen from 2007.

23, TITLE:      TabFact: A Large-scale Dataset for Table-based Fact Verification
http://arxiv.org/abs/1909.02164
AUTHORS:        Wenhu Chen ; Hongmin Wang ; Jianshu Chen ; Yunkai Zhang ; Hong Wang ; Shiyang Li ; Xiyou Zhou ; William Yang Wang
COMMENTS:       Accepted to ICLR 2020, 17 pages, 15 figures. Main paper has 9 pages
HIGHLIGHT:      The problem of verifying whether a textual hypothesis holds based on the given evidence, also known as fact verification, plays an important role in the study of natural language understanding and semantic representation. To this end, we construct a large-scale dataset called TabFact with 16k Wikipedia tables as the evidence for 118k human-annotated natural language statements, which are labeled as either ENTAILED or REFUTED.

24, TITLE:      Learning to Observe: Approximating Human Perceptual Thresholds for Detection of Suprathreshold Image Transformations
http://arxiv.org/abs/1912.06433
AUTHORS:        Alan Dolhasz ; Carlo Harvey ; Ian Williams
COMMENTS:       8 pages + references
HIGHLIGHT:      In this paper, we propose to directly approximate the perceptual function performed by human observers completing a visual detection task.

25, TITLE:      Intensionality, Intensional Recursion, and the Gödel-Löb axiom
http://arxiv.org/abs/1703.01288
AUTHORS:        G. A. Kavvos
COMMENTS:       Presented at IMLA 2017. Revised version following post-conference review
HIGHLIGHT:      Intensionality, Intensional Recursion, and the Gödel-Löb axiom

26, TITLE:      ML-driven Malware that Targets AV Safety
http://arxiv.org/abs/2004.13004
AUTHORS:        Saurabh Jha ; Shengkun Cui ; Subho S. Banerjee ; Timothy Tsai ; Zbigniew Kalbarczyk ; Ravi Iyer
COMMENTS:       Accepted for DSN 2020
HIGHLIGHT:      In this paper, we introduce an attack model, a method to deploy the attack in the form of smart malware, and an experimental evaluation of its impact on production-grade autonomous driving software.

27, TITLE:      Iterative Effect-Size Bias in Ridehailing: Measuring Social Bias in Dynamic Pricing of 100 Million Rides
http://arxiv.org/abs/2006.04599
AUTHORS:        Akshat Pandey ; Aylin Caliskan
COMMENTS:       16 pages, 5 tables, 4 figures
HIGHLIGHT:      In this work we develop a random-effects based metric for the analysis of social bias in supervised machine learning prediction models where model outputs depend on U.S. locations.

28, TITLE:      Toward Adversarial Robustness via Semi-supervised Robust Training
http://arxiv.org/abs/2003.06974
AUTHORS:        Yiming Li ; Baoyuan Wu ; Yan Feng ; Yanbo Fan ; Yong Jiang ; Zhifeng Li ; Shutao Xia
COMMENTS:       19 pages
HIGHLIGHT:      In this work, we propose a novel defense method, the robust training (RT), by jointly minimizing two separated risks ($R_{stand}$ and $R_{rob}$), which is with respect to the benign example and its neighborhoods respectively.

29, TITLE:      Deep Back-Projection Networks for Single Image Super-resolution
http://arxiv.org/abs/1904.05677
AUTHORS:        Muhammad Haris ; Greg Shakhnarovich ; Norimichi Ukita
COMMENTS:       To appear in TPAMI 2020. The code is available at https://github.com/alterzero/DBPN-Pytorch arXiv admin note: substantial text overlap with arXiv:1803.02735
HIGHLIGHT:      We propose Deep Back-Projection Networks (DBPN), the winner of two image super-resolution challenges (NTIRE2018 and PIRM2018), that exploit iterative up- and down-sampling layers.

30, TITLE:      Real-time Human Activity Recognition Using Conditionally Parametrized Convolutions on Mobile and Wearable Devices
http://arxiv.org/abs/2006.03259
AUTHORS:        Xin Cheng ; Lei Zhang ; Yin Tang ; Yue Liu ; Hao Wu ; Jun He
COMMENTS:       10 pages,14 figures
HIGHLIGHT:      In this paper, we for the first time propose a computation efficient CNN using conditionally parametrized convolution for real-time HAR on mobile and wearable devices.

31, TITLE:      G3AN: Disentangling Appearance and Motion for Video Generation
http://arxiv.org/abs/1912.05523
AUTHORS:        Yaohui Wang ; Piotr Bilinski ; Francois Bremond ; Antitza Dantcheva
COMMENTS:       CVPR 2020, project link https://wyhsirius.github.io/G3AN/
HIGHLIGHT:      To tackle this challenge, we introduce G$^{3}$AN, a novel spatio-temporal generative model, which seeks to capture the distribution of high dimensional video data and to model appearance and motion in disentangled manner.

32, TITLE:      Ansor : Generating High-Performance Tensor Programs for Deep Learning
http://arxiv.org/abs/2006.06762
AUTHORS:        Lianmin Zheng ; Chengfan Jia ; Minmin Sun ; Zhao Wu ; Cody Hao Yu ; Ameer Haj-Ali ; Yida Wang ; Jun Yang ; Danyang Zhuo ; Koushik Sen ; Joseph E. Gonzalez ; Ion Stoica
HIGHLIGHT:      We present Ansor, a tensor program generation framework for deep learning applications.

33, TITLE:      FP-Stereo: Hardware-Efficient Stereo Vision for Embedded Applications
http://arxiv.org/abs/2006.03250
AUTHORS:        Jieru Zhao ; Tingyuan Liang ; Liang Feng ; Wenchao Ding ; Sharad Sinha ; Wei Zhang ; Shaojie Shen
COMMENTS:       IEEE International Conference on Field Programmable Logic and Applications (FPL), 2020
HIGHLIGHT:      To reduce the design effort and achieve the right balance, we propose FP-Stereo for building high-performance stereo matching pipelines on FPGAs automatically.

34, TITLE:      Adversarial Attacks and Defense on Texts: A Survey
http://arxiv.org/abs/2005.14108
AUTHORS:        Aminul Huq ; Mst. Tasnim Pervin
HIGHLIGHT:      In this manuscript, we accumulated and analyzed different attacking techniques and various defense models to provide a more comprehensive idea.

35, TITLE:      Fingerprint Synthesis: Search with 100 Million Prints
http://arxiv.org/abs/1912.07195
AUTHORS:        Vishesh Mistry ; Joshua J. Engelsma ; Anil K. Jain
HIGHLIGHT:      To address this problem, we utilize a Generative Adversarial Network (GAN) to synthesize a fingerprint dataset consisting of 100 million fingerprint images.

36, TITLE:      UnrealText: Synthesizing Realistic Scene Text Images from the Unreal World
http://arxiv.org/abs/2003.10608
AUTHORS:        Shangbang Long ; Cong Yao
COMMENTS:       Accepted to CVPR 2020 (updated citation and comparison with a previous work)
HIGHLIGHT:      In this paper, we introduce UnrealText, an efficient image synthesis method that renders realistic images via a 3D graphics engine.

37, TITLE:      Stochastic Inverse Reinforcement Learning
http://arxiv.org/abs/1905.08513
AUTHORS:        Ce Ju ; Dong Eui Chang
COMMENTS:       8+1 pages, 5 figures, Accepted for WorkShop of NeurIPS 2019
HIGHLIGHT:      In this paper, we generalize the original IRL problem to recover a probability distribution for reward functions.

38, TITLE:      Text Complexity Classification Based on Linguistic Information: Application to Intelligent Tutoring of ESL
http://arxiv.org/abs/2001.01863
AUTHORS:        M. Zakaria Kurdi
COMMENTS:       This is an unpublished pre-print, the JDMDH journal requires submission to arxiv.org before the submission to the journal (see the link: https://jdmdh.episciences.org/page/submissions#)
HIGHLIGHT:      The goal of this work is to build a classifier that can identify text complexity within the context of teaching reading to English as a Second Language (ESL) learners.

39, TITLE:      Deep Reflection Prior
http://arxiv.org/abs/1912.03623
AUTHORS:        Yingda Yin ; Qingnan Fan ; Dongdong Chen ; Yujie Wang ; Angelica Aviles-Rivero ; Ruoteng Li ; Carola-Bibiane Schnlieb ; Dani Lischinski ; Baoquan Chen
HIGHLIGHT:      In this paper, we propose a learning-based approach that captures the reflection statistical prior for single image reflection removal.

40, TITLE:      On the Expressivity of Neural Networks for Deep Reinforcement Learning
http://arxiv.org/abs/1910.05927
AUTHORS:        Kefan Dong ; Yuping Luo ; Tengyu Ma
COMMENTS:       Accepted in ICML 2020. Title of previous version was "Bootstrapping the Expressivity with Model-based Planning"
HIGHLIGHT:      We compare the model-free reinforcement learning with the model-based approaches through the lens of the expressive power of neural networks for policies, $Q$-functions, and dynamics.

41, TITLE:      S4NN: temporal backpropagation for spiking neural networks with one spike per neuron
http://arxiv.org/abs/1910.09495
AUTHORS:        Saeed Reza Kheradpisheh ; Timothée Masquelier
HIGHLIGHT:      We propose a new supervised learning rule for multilayer spiking neural networks (SNNs) that use a form of temporal coding known as rank-order-coding.

42, TITLE:      MSG-GAN: Multi-Scale Gradients for Generative Adversarial Networks
http://arxiv.org/abs/1903.06048
AUTHORS:        Animesh Karnewar ; Oliver Wang
COMMENTS:       CVPR 2020 (Main Conference). Work sponsored by TomTom and Adobe. Code repository: https://github.com/akanimax/msg-stylegan-tf
HIGHLIGHT:      In this work, we propose the Multi-Scale Gradient Generative Adversarial Network (MSG-GAN), a simple but effective technique for addressing this by allowing the flow of gradients from the discriminator to the generator at multiple scales.

43, TITLE:      Least squares binary quantization of neural networks
http://arxiv.org/abs/2001.02786
AUTHORS:        Hadi Pouransari ; Zhucheng Tu ; Oncel Tuzel
HIGHLIGHT:      In this work, we focus on the binary quantization, in which values are mapped to -1 and 1.

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

45, TITLE:      Generating Diverse and Consistent QA pairs from Contexts with Information-Maximizing Hierarchical Conditional VAEs
http://arxiv.org/abs/2005.13837
AUTHORS:        Dong Bok Lee ; Seanie Lee ; Woo Tae Jeong ; Donghwan Kim ; Sung Ju Hwang
COMMENTS:       ACL 2020
HIGHLIGHT:      In this work, we propose a hierarchical conditional variational autoencoder (HCVAE) for generating QA pairs given unstructured texts as contexts, while maximizing the mutual information between generated QA pairs to ensure their consistency.

46, TITLE:      GLAMpoints: Greedily Learned Accurate Match points
http://arxiv.org/abs/1908.06812
AUTHORS:        Prune Truong ; Stefanos Apostolopoulos ; Agata Mosinska ; Samuel Stucky ; Carlos Ciller ; Sandro De Zanet
HIGHLIGHT:      We introduce a novel CNN-based feature point detector - GLAMpoints - learned in a semi-supervised manner.

47, TITLE:      Systematically designing better instance counting models on cell images with Neural Arithmetic Logic Units
http://arxiv.org/abs/2004.06674
AUTHORS:        Ashish Rana ; Taranveer Singh ; Harpreet Singh ; Neeraj Kumar ; Prashant Singh Rana
COMMENTS:       * code repository for project: https://github.com/ashishrana160796/nalu-cell-counting
HIGHLIGHT:      By making better predictions for higher ranges of cell count we are aiming to create better generalization systems for cell counting.

48, TITLE:      Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifiers
http://arxiv.org/abs/1912.03277
AUTHORS:        Divyat Mahajan ; Chenhao Tan ; Amit Sharma
COMMENTS:       2019 NeurIPS Workshop on Do the right thing: Machine learning and Causal Inference for improved decision making
HIGHLIGHT:      We formulate the problem of feasibility as preserving causal relationships among input features and present a method that uses (partial) structural causal models to generate actionable counterfactuals.

49, TITLE:      A closed-form solution to estimate uncertainty in non-rigid structure from motion
http://arxiv.org/abs/2005.04810
AUTHORS:        Jingwei Song ; Mitesh Patel
COMMENTS:       9 pages, 2 figures
HIGHLIGHT:      In this paper, we present a statistical inference on the element-wise uncertainty quantification of the estimated deforming 3D shape points in the case of the exact low-rank SDP problem.

50, TITLE:      Temporal Pyramid Network for Action Recognition
http://arxiv.org/abs/2004.03548
AUTHORS:        Ceyuan Yang ; Yinghao Xu ; Jianping Shi ; Bo Dai ; Bolei Zhou
COMMENTS:       To appear in CVPR 2020. Code is available at https://github.com/decisionforce/TPN
HIGHLIGHT:      In this work we propose a generic Temporal Pyramid Network (TPN) at the feature-level, which can be flexibly integrated into 2D or 3D backbone networks in a plug-and-play manner.

51, TITLE:      Deep covariate-learning: optimising information extraction from terrain texture for geostatistical modelling applications
http://arxiv.org/abs/2005.11194
AUTHORS:        Charlie Kirkwood
COMMENTS:       14 pages, 8 figures, submitted to journal
HIGHLIGHT:      In this paper we present a solution to this problem in the form of a deep learning approach to automatically deriving optimal task-specific terrain texture covariates from a standard SRTM 90m gridded digital elevation model (DEM).

52, TITLE:      Synthetic Learning: Learn From Distributed Asynchronized Discriminator GAN Without Sharing Medical Image Data
http://arxiv.org/abs/2006.00080
AUTHORS:        Qi Chang ; Hui Qu ; Yikai Zhang ; Mert Sabuncu ; Chao Chen ; Tong Zhang ; Dimitris Metaxas
HIGHLIGHT:      In this paper, we propose a data privacy-preserving and communication efficient distributed GAN learning framework named Distributed Asynchronized Discriminator GAN (AsynDGAN).

53, TITLE:      Fast and Automatic Periacetabular Osteotomy Fragment Pose Estimation Using Intraoperatively Implanted Fiducials and Single-View Fluoroscopy
http://arxiv.org/abs/1910.10187
AUTHORS:        Robert Grupp ; Ryan Murphy ; Rachel Hegeman ; Clayton Alexander ; Mathias Unberath ; Yoshito Otake ; Benjamin McArthur ; Mehran Armand ; Russell Taylor
COMMENTS:       Revised article to address reviewer comments. Under review for Physics in Medicine and Biology. Supplementary video at https://youtu.be/0E0U9G81q8g
HIGHLIGHT:      We propose a computer assisted approach that uses a single fluoroscopic view and quickly reports the pose of an acetabular fragment without any user input or initialization.

54, TITLE:      Dance Revolution: Long Sequence Dance Generation with Music via Curriculum Learning
http://arxiv.org/abs/2006.06119
AUTHORS:        Ruozi Huang ; Huang Hu ; Wei Wu ; Kei Sawada ; Mi Zhang
COMMENTS:       Submitted to NeurIPS 2020
HIGHLIGHT:      To further alleviate the error accumulation in human motion synthesis, we introduce a dynamic auto-condition training strategy as a new curriculum learning method to facilitate the long-term dance generation.

55, TITLE:      FastReID: A Pytorch Toolbox for General Instance Re-identification
http://arxiv.org/abs/2006.02631
AUTHORS:        Lingxiao He ; Xingyu Liao ; Wu Liu ; Xinchen Liu ; Peng Cheng ; Tao Mei
HIGHLIGHT:      We present FastReID, as a widely used general instance re-identification (re-id) software system in JD AI Research, has been applied to video analysis such as person re-id, vehicle re-id and image retrieval.

56, TITLE:      Featherweight Go
http://arxiv.org/abs/2005.11710
AUTHORS:        Robert Griesemer ; Raymond Hu ; Wen Kokke ; Julien Lange ; Ian Lance Taylor ; Bernardo Toninho ; Philip Wadler ; Nobuko Yoshida
COMMENTS:       a full version
HIGHLIGHT:      We describe a design for generics in Go inspired by previous work on Featherweight Java by Igarashi, Pierce, and Wadler.

57, TITLE:      How Far are We from Effective Context Modeling? An Exploratory Study on Semantic Parsing in Context
http://arxiv.org/abs/2002.00652
AUTHORS:        Qian Liu ; Bei Chen ; Jiaqi Guo ; Jian-Guang Lou ; Bin Zhou ; Dongmei Zhang
COMMENTS:       Accepted by IJCAI2020 (http://static.ijcai.org/2020-accepted_papers.html). SOLE copyright holder is IJCAI (International Joint Conferences on Artificial Intelligence), all rights reserved
HIGHLIGHT:      We present a grammar-based decoding semantic parser and adapt typical context modeling methods on top of it.

58, TITLE:      A Linear Algebra Approach for Detecting Binomiality of Steady State Ideals of Reversible Chemical Reaction Networks
http://arxiv.org/abs/2002.12693
AUTHORS:        Hamid Rahkooy ; Ovidiu Radulescu ; Thomas Sturm
HIGHLIGHT:      On these grounds we propose an efficient algorithm for detecting unconditional binomiality.

59, TITLE:      Investigating Task-driven Latent Feasibility for Nonconvex Image Modeling
http://arxiv.org/abs/1910.08242
AUTHORS:        Risheng Liu ; Pan Mu ; Jian Chen ; Xin Fan ; Zhongxuan Luo
COMMENTS:       11 pages
HIGHLIGHT:      In this work, we provide a new perspective, named Task-driven Latent Feasibility (TLF), to incorporate specific task information to narrow down the solution space for the optimization-based image modeling problem.

60, TITLE:      Distilling Effective Supervision from Severe Label Noise
http://arxiv.org/abs/1910.00701
AUTHORS:        Zizhao Zhang ; Han Zhang ; Sercan O. Arik ; Honglak Lee ; Tomas Pfister
COMMENTS:       CVPR2020
HIGHLIGHT:      We present a holistic framework to train deep neural networks in a way that is highly invulnerable to label noise.

61, TITLE:      Circle Loss: A Unified Perspective of Pair Similarity Optimization
http://arxiv.org/abs/2002.10857
AUTHORS:        Yifan Sun ; Changmao Cheng ; Yuhan Zhang ; Chi Zhang ; Liang Zheng ; Zhongdao Wang ; Yichen Wei
HIGHLIGHT:      This paper provides a pair similarity optimization viewpoint on deep feature learning, aiming to maximize the within-class similarity $s_p$ and minimize the between-class similarity $s_n$.

62, TITLE:      A Survey of Model Compression and Acceleration for Deep Neural Networks
http://arxiv.org/abs/1710.09282
AUTHORS:        Yu Cheng ; Duo Wang ; Pan Zhou ; Tao Zhang
COMMENTS:       Published in IEEE Signal Processing Magazine, updated version including more recent works
HIGHLIGHT:      In this paper, we review the recent techniques for compacting and accelerating DNN models.

63, TITLE:      Tree Stack Memory Units
http://arxiv.org/abs/1911.01545
AUTHORS:        Forough Arabshahi ; Zhichu Lu ; Sameer Singh ; Animashree Anandkumar
HIGHLIGHT:      To overcome this, we propose Tree Stack Memory Units (Tree-SMUs), a novel memory augmented recursive neural network whose nodes consist of a differentiable stack.

64, TITLE:      Bridging Anaphora Resolution as Question Answering
http://arxiv.org/abs/2004.07898
AUTHORS:        Yufang Hou
COMMENTS:       accepted at ACL2020
HIGHLIGHT:      In this paper, we cast bridging anaphora resolution as question answering based on context.

65, TITLE:      CausaLM: Causal Model Explanation Through Counterfactual Language Models
http://arxiv.org/abs/2005.13407
AUTHORS:        Amir Feder ; Nadav Oved ; Uri Shalit ; Roi Reichart
COMMENTS:       Our code and data are available at: https://amirfeder.github.io/CausaLM/ Under review for the Computational Linguistics journal
HIGHLIGHT:      To bridge that gap, we propose CausaLM, a framework for producing causal model explanations using counterfactual language representation models.

66, TITLE:      EfficientDet: Scalable and Efficient Object Detection
http://arxiv.org/abs/1911.09070
AUTHORS:        Mingxing Tan ; Ruoming Pang ; Quoc V. Le
COMMENTS:       CVPR 2020
HIGHLIGHT:      In this paper, we systematically study neural network architecture design choices for object detection and propose several key optimizations to improve efficiency.

67, TITLE:      3D-MiniNet: Learning a 2D Representation from Point Clouds for Fast and Efficient 3D LIDAR Semantic Segmentation
http://arxiv.org/abs/2002.10893
AUTHORS:        Iñigo Alonso ; Luis Riazuelo ; Luis Montesano ; Ana C. Murillo
COMMENTS:       8 pages, 4 figures
HIGHLIGHT:      This work presents 3D-MiniNet, a novel approach for LIDAR semantic segmentation that combines 3D and 2D learning layers.

68, TITLE:      SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving
http://arxiv.org/abs/2003.03653
AUTHORS:        Tiago Cortinhal ; George Tzelepis ; Eren Erdal Aksoy
HIGHLIGHT:      In this paper, we introduce SalsaNext for the uncertainty-aware semantic segmentation of a full 3D LiDAR point cloud in real-time.

69, TITLE:      Machine learning as a model for cultural learning: Teaching an algorithm what it means to be fat
http://arxiv.org/abs/2003.12133
AUTHORS:        Alina Arseniev-Koehler ; Jacob G. Foster
HIGHLIGHT:      In this paper, we provide a theoretical account of such cultural learning.

70, TITLE:      The Average-Case Time Complexity of Certifying the Restricted Isometry Property
http://arxiv.org/abs/2005.11270
AUTHORS:        Yunzi Ding ; Dmitriy Kunisky ; Alexander S. Wein ; Afonso S. Bandeira
COMMENTS:       12 pages
HIGHLIGHT:      In this paper, we investigate the exact average-case time complexity of certifying the RIP property for $M\times N$ matrices with i.i.d. $\mathcal{N}(0,1/M)$ entries, in the "possible but hard" regime $\sqrt{M} \ll s\lesssim M/\log N$, assuming that $M$ scales proportional to $N$.

71, TITLE:      Learning a Unified Sample Weighting Network for Object Detection
http://arxiv.org/abs/2006.06568
AUTHORS:        Qi Cai ; Yingwei Pan ; Yu Wang ; Jingen Liu ; Ting Yao ; Tao Mei
COMMENTS:       CVPR 2020; The source code and model are publicly available at: \url{https://github.com/caiqi/sample-weighting-network}
HIGHLIGHT:      To this end, we devise a general loss function to cover most region-based object detectors with various sampling strategies, and then based on it we propose a unified sample weighting network to predict a sample's task weights.

72, TITLE:      An Efficient Accelerator Design Methodology for Deformable Convolutional Networks
http://arxiv.org/abs/2006.05238
AUTHORS:        Saehyun Ahn ; Jung-Woo Chang ; Suk-Ju Kang
COMMENTS:       IEEE International Conference on Image Processing (ICIP) 2020
HIGHLIGHT:      In this paper, we present a novel approach to accelerate deformable convolution on FPGA.

73, TITLE:      Hardness of Minimum Barrier Shrinkage and Minimum Installation Path
http://arxiv.org/abs/1910.04228
AUTHORS:        Sergio Cabello ; Éric Colin de Verdière
HIGHLIGHT:      The task is to shrink the disks, each one possibly by a different amount, so that we can draw an $x$-$y$ curve that is disjoint from the interior of the shrunken disks, and the sum of the decreases in the radii is minimized.

74, TITLE:      Rethinking Localization Map: Towards Accurate Object Perception with Self-Enhancement Maps
http://arxiv.org/abs/2006.05220
AUTHORS:        Xiaolin Zhang ; Yunchao Wei ; Yi Yang ; Fei Wu
HIGHLIGHT:      We propose a two-stage approach to generate the localization maps by simply comparing the similarity of point-wise features between the high-activation and the rest pixels.

75, TITLE:      H3DNet: 3D Object Detection Using Hybrid Geometric Primitives
http://arxiv.org/abs/2006.05682
AUTHORS:        Zaiwei Zhang ; Bo Sun ; Haitao Yang ; Qixing Huang
HIGHLIGHT:      We introduce H3DNet, which takes a colorless 3D point cloud as input and outputs a collection of oriented object bounding boxes (or BB) and their semantic labels.

76, TITLE:      DeepGCNs: Making GCNs Go as Deep as CNNs
http://arxiv.org/abs/1910.06849
AUTHORS:        Guohao Li ; Matthias Müller ; Guocheng Qian ; Itzel C. Delgadillo ; Abdulellah Abualshour ; Ali Thabet ; Bernard Ghanem
COMMENTS:       First two authors contributed equally. This work is a journal extension of our ICCV'19 paper arXiv:1904.03751
HIGHLIGHT:      We believe that the insights in this work will open avenues for future research on GCNs and their application to further tasks not explored in this paper.

77, TITLE:      STAViS: Spatio-Temporal AudioVisual Saliency Network
http://arxiv.org/abs/2001.03063
AUTHORS:        Antigoni Tsiami ; Petros Koutras ; Petros Maragos
COMMENTS:       CVPR 2020. Project page: https://github.com/atsiami/STAViS
HIGHLIGHT:      We introduce STAViS, a spatio-temporal audiovisual saliency network that combines spatio-temporal visual and auditory information in order to efficiently address the problem of saliency estimation in videos.

78, TITLE:      Dynamic Graph Message Passing Networks
http://arxiv.org/abs/1908.06955
AUTHORS:        Li Zhang ; Dan Xu ; Anurag Arnab ; Philip H. S. Torr
COMMENTS:       CVPR 2020 Oral
HIGHLIGHT:      We propose a dynamic graph message passing network, based on the message passing neural network framework, that significantly reduces the computational complexity compared to related works modelling a fully-connected graph.

79, TITLE:      Quantum Robust Fitting
http://arxiv.org/abs/2006.06986
AUTHORS:        Tat-Jun Chin ; David Suter ; Shin-Fang Chng ; James Quach
HIGHLIGHT:      In this paper, we explore the usage of quantum computers for robust fitting.

80, TITLE:      Non-Negative Bregman Divergence Minimization for Deep Direct Density Ratio Estimation
http://arxiv.org/abs/2006.06979
AUTHORS:        Masahiro Kato ; Takeshi Teshima
HIGHLIGHT:      In this paper, we introduce a non-negative correction for empirical risk using only the prior knowledge of the upper bound of the density ratio.

81, TITLE:      DeepCMB: Lensing Reconstruction of the Cosmic Microwave Background with Deep Neural Networks
http://arxiv.org/abs/1810.01483
AUTHORS:        João Caldeira ; W. L. Kimmy Wu ; Brian Nord ; Camille Avestruz ; Shubhendu Trivedi ; Kyle T. Story
COMMENTS:       19 pages; LaTeX; 12 figures; changes to match published version
HIGHLIGHT:      In this work, we demonstrate reconstruction of the CMB lensing potential with deep convolutional neural networks (CNN) - ie, a ResUNet.

82, TITLE:      Counterfactual VQA: A Cause-Effect Look at Language Bias
http://arxiv.org/abs/2006.04315
AUTHORS:        Yulei Niu ; Kaihua Tang ; Hanwang Zhang ; Zhiwu Lu ; Xian-Sheng Hua ; Ji-Rong Wen
HIGHLIGHT:      In this paper, we propose a novel cause-effect look at the language bias, where the bias is formulated as the direct effect of question on answer from the view of causal inference.

83, TITLE:      Highway Transformer: Self-Gating Enhanced Self-Attentive Networks
http://arxiv.org/abs/2004.08178
AUTHORS:        Yekun Chai ; Jin Shuo ; Xinwen Hou
COMMENTS:       Accepted at ACL 2020
HIGHLIGHT:      Through a pseudo information highway, we introduce a gated component self-dependency units (SDU) that incorporates LSTM-styled gating units to replenish internal semantic importance within the multi-dimensional latent space of individual representations.

84, TITLE:      Multiview Neural Surface Reconstruction with Implicit Lighting and Material
http://arxiv.org/abs/2003.09852
AUTHORS:        Lior Yariv ; Yoni Kasten ; Dror Moran ; Meirav Galun ; Matan Atzmon ; Ronen Basri ; Yaron Lipman
HIGHLIGHT:      In this work we address the challenging problem of multiview 3D surface reconstruction.

85, TITLE:      C3VQG: Category Consistent Cyclic Visual Question Generation
http://arxiv.org/abs/2005.07771
AUTHORS:        Shagun Uppal ; Anish Madan ; Sarthak Bhagat ; Yi Yu ; Rajiv Ratn Shah
HIGHLIGHT:      In this paper, we try to exploit the different visual cues and concepts in an image to generate questions using a variational autoencoder (VAE) without ground-truth answers.

86, TITLE:      Robust Baggage Detection and Classification Based on Local Tri-directional Pattern
http://arxiv.org/abs/2006.07345
AUTHORS:         Shahbano ; Muhammad Abdullah ; Kashif Inayat
HIGHLIGHT:      Therefore, to overcome these shortcomings, our research proposed a detection algorithm for a human with or without carrying baggage.

87, TITLE:      Intelligent Coordination among Multiple Traffic Intersections Using Multi-Agent Reinforcement Learning
http://arxiv.org/abs/1912.03851
AUTHORS:        Ujwal Padam Tewari ; Vishal Bidawatka ; Varsha Raveendran ; Vinay Sudhakaran ; Shreedhar Kodate Shreeshai ; Jayanth V. Kulkarni
COMMENTS:       Accepted in the NeurIPS 2019 Deep RL Workshop : https://sites.google.com/view/deep-rl-workshop-neurips-2019/home
HIGHLIGHT:      We use Asynchronous Advantage Actor Critic (A3C) for implementing an AI agent in the controllers that optimize flow of traffic across a single intersection and then extend it to multiple intersections by considering a multi-agent setting.

88, TITLE:      Feature Extraction for Hyperspectral Imagery: The Evolution from Shallow to Deep
http://arxiv.org/abs/2003.02822
AUTHORS:        Behnood Rasti ; Danfeng Hong ; Renlong Hang ; Pedram Ghamisi ; Xudong Kang ; Jocelyn Chanussot ; Jon Atli Benediktsson
HIGHLIGHT:      In more detail, this paper provides a bird's eye view over shallow (both supervised and unsupervised) and deep feature extraction approaches specifically dedicated to the topic of hyperspectral feature extraction and its application on hyperspectral image classification.

89, TITLE:      Global Voices: Crossing Borders in Automatic News Summarization
http://arxiv.org/abs/1910.00421
AUTHORS:        Khanh Nguyen ; Hal Daumé III
COMMENTS:       NewSum workshop at EMNLP 2019, 8 pages
HIGHLIGHT:      We construct Global Voices, a multilingual dataset for evaluating cross-lingual summarization methods.

90, TITLE:      Incorporating Uncertain Segmentation Information into Chinese NER for Social Media Text
http://arxiv.org/abs/2004.06384
AUTHORS:        Shengbin Jia ; Ling Ding ; Xiaojun Chen ; Shijia E ; Yang Xiang
COMMENTS:       SocialNLP@ACL2020
HIGHLIGHT:      In this paper, we propose a model (UIcwsNN) that specializes in identifying entities from Chinese social media text, especially by leveraging ambiguous information of word segmentation.

91, TITLE:      Gradient Boosted Normalizing Flows
http://arxiv.org/abs/2002.11896
AUTHORS:        Robert Giaquinto ; Arindam Banerjee
COMMENTS:       Under review
HIGHLIGHT:      We propose an alternative: Gradient Boosted Normalizing Flows (GBNF) model a density by successively adding new NF components with gradient boosting.

92, TITLE:      Span-based Localizing Network for Natural Language Video Localization
http://arxiv.org/abs/2004.13931
AUTHORS:        Hao Zhang ; Aixin Sun ; Wei Jing ; Joey Tianyi Zhou
COMMENTS:       To appear at ACL 2020
HIGHLIGHT:      In this work, we address NLVL task with a span-based QA approach by treating the input video as text passage.

93, TITLE:      Learning scale-variant features for robust iris authentication with deep learning based ensemble framework
http://arxiv.org/abs/1912.00756
AUTHORS:        Siming Zheng ; Rahmita Wirza O. K. Rahmat ; Fatimah Khalid ; Nurul Amelina Nasharuddin
COMMENTS:       This is the second revision for updating the formation, logical and image captions
HIGHLIGHT:      Importantly, the proposed model can be trained using the different spectrum of samples, such as Visible Wavelength (VW) and Near Infrared (NIR) iris biometric databases.

94, TITLE:      Fine-grained Image-to-Image Transformation towards Visual Recognition
http://arxiv.org/abs/2001.03856
AUTHORS:        Wei Xiong ; Yutong He ; Yixuan Zhang ; Wenhan Luo ; Lin Ma ; Jiebo Luo
COMMENTS:       CVPR 2020 Camera Ready version
HIGHLIGHT:      In this paper, we aim at transforming an image with a fine-grained category to synthesize new images that preserve the identity of the input image, which can thereby benefit the subsequent fine-grained image recognition and few-shot learning tasks.

95, TITLE:      Algebraic Global Gadgetry for Surjective Constraint Satisfaction
http://arxiv.org/abs/2005.11307
AUTHORS:        Hubie Chen
HIGHLIGHT:      We present an algebraic framework for proving hardness results on surjective CSPs; essentially, this framework computes global gadgetry that permits one to present a reduction from a classical CSP to a surjective CSP.

96, TITLE:      Seeing the World in a Bag of Chips
http://arxiv.org/abs/2001.04642
AUTHORS:        Jeong Joon Park ; Aleksander Holynski ; Steve Seitz
COMMENTS:       CVPR 2020
HIGHLIGHT:      Our contributions include 1) modeling highly specular objects, 2) modeling inter-reflections and Fresnel effects, and 3) enabling surface light field reconstruction with the same input needed to reconstruct shape alone.

97, TITLE:      A Review of Cooperative Multi-Agent Deep Reinforcement Learning
http://arxiv.org/abs/1908.03963
AUTHORS:        Afshin OroojlooyJadid ; Davood Hajinezhad
HIGHLIGHT:      In this review article, we have mostly focused on recent papers on Multi-Agent Reinforcement Learning (MARL) than the older papers, unless it was necessary.

98, TITLE:      Integrating global spatial features in CNN based Hyperspectral/SAR imagery classification
http://arxiv.org/abs/2006.00234
AUTHORS:        Fan Zhang ; MinChao Yan ; Chen Hu ; Jun Ni ; Fei Ma
HIGHLIGHT:      This paper proposed a novel method to take into the information of remote sensing image, i.e., geographic latitude-longitude information.

99, TITLE:      A category theoretical argument for causal inference
http://arxiv.org/abs/2004.09999
AUTHORS:        Rémy Tuyéras
COMMENTS:       48 pages, 1 figure; v2: typos corrected and clarifications linking the results added
HIGHLIGHT:      The goal of this paper is to design a causal inference method accounting for complex interactions between causal factors.

100, TITLE:     Contextualizing Hate Speech Classifiers with Post-hoc Explanation
http://arxiv.org/abs/2005.02439
AUTHORS:        Brendan Kennedy ; Xisen Jin ; Aida Mostafazadeh Davani ; Morteza Dehghani ; Xiang Ren
COMMENTS:       To appear in Proceedings of the 2020 Annual Conference of the Association for Computational Linguistics
HIGHLIGHT:      Our approach improved over baselines in limiting false positives on out-of-domain data while maintaining or improving in-domain performance.

101, TITLE:     BabyWalk: Going Farther in Vision-and-Language Navigation by Taking Baby Steps
http://arxiv.org/abs/2005.04625
AUTHORS:        Wang Zhu ; Hexiang Hu ; Jiacheng Chen ; Zhiwei Deng ; Vihan Jain ; Eugene Ie ; Fei Sha
COMMENTS:       Accepted by ACL 2020
HIGHLIGHT:      In this paper, we study how an agent can navigate long paths when learning from a corpus that consists of shorter ones. We create two new benchmark datasets (of long navigation tasks) and use them in conjunction with existing ones to examine BabyWalk's generalization ability.

102, TITLE:     Dual Learning for Semi-Supervised Natural Language Understanding
http://arxiv.org/abs/2004.12299
AUTHORS:        Su Zhu ; Ruisheng Cao ; Kai Yu
COMMENTS:       12 pages, 4 figures; Accepted for IEEE/ACM Transactions on Audio Speech and Language Processing
HIGHLIGHT:      In this work, we introduce a dual task of NLU, semantic-to-sentence generation (SSG), and propose a new framework for semi-supervised NLU with the corresponding dual model.

103, TITLE:     Can AI help in screening Viral and COVID-19 pneumonia?
http://arxiv.org/abs/2003.13145
AUTHORS:        Muhammad E. H. Chowdhury ; Tawsifur Rahman ; Amith Khandakar ; Rashid Mazhar ; Muhammad Abdul Kadir ; Zaid Bin Mahbub ; Khandaker Reajul Islam ; Muhammad Salman Khan ; Atif Iqbal ; Nasser Al-Emadi ; Mamun Bin Ibne Reaz ; T. I. Islam
COMMENTS:       12 pages, 9 Figures
HIGHLIGHT:      The aim of this paper is to propose a robust technique for automatic detection of COVID-19 pneumonia from digital chest X-ray images applying pre-trained deep-learning algorithms while maximizing the detection accuracy.

104, TITLE:     Resurrecting Submodularity for Neural Text Generation
http://arxiv.org/abs/1911.03014
AUTHORS:        Simeng Han ; Xiang Lin ; Shafiq Joty
HIGHLIGHT:      We propose diminishing attentions, a class of novel attention mechanisms that exploit the properties of submodular functions.

105, TITLE:     Flickering Adversarial Attacks against Video Recognition Networks
http://arxiv.org/abs/2002.05123
AUTHORS:        Itay Naeh ; Roi Pony ; Shie Mannor
COMMENTS:       10 pages, 5 figures
HIGHLIGHT:      In this work we present a manipulation scheme for fooling video classifiers by introducing a flickering temporal perturbation that is practically unnoticeable by human observers and is implementable in the real world.

106, TITLE:     A Review of Emergency Incident Prediction, Resource Allocation and Dispatch Models
http://arxiv.org/abs/2006.04200
AUTHORS:        Ayan Mukhopadhyay ; Geoffrey Pettet ; Sayyed Vazirizade ; Yevgeniy Vorobeychik ; Mykel Kochenderfer ; Abhishek Dubey
HIGHLIGHT:      In this survey, we present models for incident prediction, resource allocation, and dispatch concerning urban emergency incidents like accidents and crimes.

107, TITLE:     Few-shot Neural Architecture Search
http://arxiv.org/abs/2006.06863
AUTHORS:        Yiyang Zhao ; Linnan Wang ; Yuandong Tian ; Rodrigo Fonseca ; Tian Guo
HIGHLIGHT:      In this work, we propose few-shot NAS that explores the choice of using multiple super-nets: each super-net is pre-trained to be in charge of a sub-region of the search space.