2020.04.14.txt

==========New Papers==========
1, TITLE:       MLR: A Two-stage Conversational Query Rewriting Model with Multi-task Learning
http://arxiv.org/abs/2004.05812
AUTHORS:        Shuangyong Song ; Chao Wang ; Qianqian Xie ; Xinxing Zu ; Huan Chen ; Haiqing Chen
HIGHLIGHT:      In this paper, we propose the conversational query rewriting model - MLR, which is a Multi-task model on sequence Labeling and query Rewriting.

2, TITLE:       Public Self-consciousness for Endowing Dialogue Agents with Consistent Persona
http://arxiv.org/abs/2004.05816
AUTHORS:        Hyunwoo Kim ; Byeongchang Kim ; Gunhee Kim
COMMENTS:       Accepted paper at ICLR 2020 BAICS workshop (Oral)
HIGHLIGHT:      Inspired by social cognition and pragmatics, we model public self-consciousness in dialogue agents through an imaginary listener to improve consistency.

3, TITLE:       MulayCap: Multi-layer Human Performance Capture Using A Monocular Video Camera
http://arxiv.org/abs/2004.05815
AUTHORS:        Zhaoqi Su ; Weilin Wan ; Tao Yu ; Lingjie Liu ; Lu Fang ; Wenping Wang ; Yebin Liu
HIGHLIGHT:      We introduce MulayCap, a novel human performance capture method using a monocular video camera without the need for pre-scanning.

4, TITLE:       Monocular Depth Estimation with Self-supervised Instance Adaptation
http://arxiv.org/abs/2004.05821
AUTHORS:        Robert McCraith ; Lukas Neumann ; Andrew Zisserman ; Andrea Vedaldi
COMMENTS:       IROS submission, 7 pages
HIGHLIGHT:      To address this mixed setting,we proposed a new approach that extends any off-the-shelfself-supervised monocular depth reconstruction system to usemore than one image at test time.

5, TITLE:       Neural Machine Translation: Challenges, Progress and Future
http://arxiv.org/abs/2004.05809
AUTHORS:        Jiajun Zhang ; Chengqing Zong
COMMENTS:       Invited Review of Science China Technological Sciences
HIGHLIGHT:      Nowadays, neural machine translation (NMT) which models direct mapping between source and target languages with deep neural networks has achieved a big breakthrough in translation performance and become the de facto paradigm of MT. This article makes a review of NMT framework, discusses the challenges in NMT, introduces some exciting recent progresses and finally looks forward to some potential future research trends.

6, TITLE:       Unified Multi-Criteria Chinese Word Segmentation with BERT
http://arxiv.org/abs/2004.05808
AUTHORS:        Zhen Ke ; Liang Shi ; Erli Meng ; Bin Wang ; Xipeng Qiu ; Xuanjing Huang
HIGHLIGHT:      In this paper, we combine the superiority of the unified framework and pretrained language model, and propose a unified MCCWS model based on BERT.

7, TITLE:       ProFormer: Towards On-Device LSH Projection Based Transformers
http://arxiv.org/abs/2004.05801
AUTHORS:        Chinnadhurai Sankar ; Sujith Ravi ; Zornitsa Kozareva
HIGHLIGHT:      To surmount these challenges, we introduce ProFormer -- a projection based transformer architecture that is faster and lighter making it suitable to deploy to memory constraint devices and preserve user privacy.

8, TITLE:       Quantifying Notes Revisited
http://arxiv.org/abs/2004.05802
AUTHORS:        Hans van Ditmarsch
HIGHLIGHT:      In this survey we review several dynamic epistemic logics with modalities representing quantification over information change.

9, TITLE:       Unsupervised Few-shot Learning via Distribution Shift-based Augmentation
http://arxiv.org/abs/2004.05805
AUTHORS:        Tiexin Qin ; Wenbin Li ; Yinghuan Shi ; Yang Gao
COMMENTS:       10 pages, 4 figures. Code: https://github.com/WonderSeven/ULDA
HIGHLIGHT:      Instead, in this paper, we develop a novel framework called \emph{Unsupervised Few-shot Learning via Distribution Shift-based Data Augmentation} (ULDA), which pays attention to the distribution diversity inside each constructed pretext few-shot task when using data augmentation.

10, TITLE:      Multi-modal Datasets for Super-resolution
http://arxiv.org/abs/2004.05804
AUTHORS:        Haoran Li ; Weihong Quan ; Meijun Yan ; Jin zhang ; Xiaoli Gong ; Jin Zhou
HIGHLIGHT:      Nowdays, most datasets used to train and evaluate super-resolution models are single-modal simulation datasets. In contrast, we first proposed real-world black-and-white old photo datasets for super-resolution (OID-RW), which is constructed using two methods of manually filling pixels and shooting with different cameras. At the same time, we also propose a multi-modal degradation dataset (MDD400) to solve the super-resolution reconstruction in real-life image degradation scenarios.

11, TITLE:      SPCNet:Spatial Preserve and Content-aware Network for Human Pose Estimation
http://arxiv.org/abs/2004.05834
AUTHORS:        Yabo Xiao ; Dongdong Yu ; Xiaojuan Wang ; Tianqi Lv ; Yiqi Fan ; Lingrui Wu
COMMENTS:       8 pages,6 figures, accepted for presentation at the 24th European Conference on Artificial Intelligence (ECAI 2020)
HIGHLIGHT:      To alleviate these issues, we propose a novel Spatial Preserve and Content-aware Network(SPCNet), which includes two effective modules: Dilated Hourglass Module(DHM) and Selective Information Module(SIM).

12, TITLE:      From Machine Reading Comprehension to Dialogue State Tracking: Bridging the Gap
http://arxiv.org/abs/2004.05827
AUTHORS:        Shuyang Gao ; Sanchit Agarwal ; Tagyoung Chung ; Di Jin ; Dilek Hakkani-Tur
HIGHLIGHT:      In this paper, we propose using machine reading comprehension (RC) in state tracking from two perspectives: model architectures and datasets.

13, TITLE:      From Inference to Generation: End-to-end Fully Self-supervised Generation of Human Face from Speech
http://arxiv.org/abs/2004.05830
AUTHORS:        Hyeong-Seok Choi ; Changdae Park ; Kyogu Lee
COMMENTS:       18 pages, 12 figures, Published as a conference paper at International Conference on Learning Representations (ICLR) 2020. (camera-ready version)
HIGHLIGHT:      To this end, we propose a multi-modal learning framework that links the inference stage and generation stage.

14, TITLE:      Exact and Approximate Algorithms for Computing a Second Hamiltonian Cycle
http://arxiv.org/abs/2004.06036
AUTHORS:        Argyrios Deligkas ; George B. Mertzios ; Paul G. Spirakis ; Viktor Zamaraev
COMMENTS:       28 pages, 4 algorithms, 5 figures
HIGHLIGHT:      In this paper we consider the following total functional problem: Given a cubic Hamiltonian graph $G$ and a Hamiltonian cycle $C_0$ of $G$, how can we compute a second Hamiltonian cycle $C_1 \neq C_0$ of $G$?

15, TITLE:      Compositional Visual Generation and Inference with Energy Based Models
http://arxiv.org/abs/2004.06030
AUTHORS:        Yilun Du ; Shuang Li ; Igor Mordatch
COMMENTS:       Website at https://energy-based-model.github.io/compositional-generation-inference/
HIGHLIGHT:      In this paper we show that energy-based models can exhibit this ability by directly combining probability distributions.

16, TITLE:      Learning from Rules Generalizing Labeled Exemplars
http://arxiv.org/abs/2004.06025
AUTHORS:        Abhijeet Awasthi ; Sabyasachi Ghosh ; Rasna Goyal ; Sunita Sarawagi
COMMENTS:       ICLR 2020 (Spotlight)
HIGHLIGHT:      We propose a rule-exemplar method for collecting human supervision to combine the efficiency of rules with the quality of instance labels.

17, TITLE:      Hamiltonian Dynamics for Real-World Shape Interpolation
http://arxiv.org/abs/2004.05199
AUTHORS:        Marvin Eisenberger ; Daniel Cremers
HIGHLIGHT:      We revisit the classical problem of 3D shape interpolation and propose a novel, physically plausible approach based on Hamiltonian dynamics.

18, TITLE:      K-spin Hamiltonian for quantum-resolvable Markov decision processes
http://arxiv.org/abs/2004.06040
AUTHORS:        Eric B. Jones ; Peter Graf ; Eliot Kapit ; Wesley Jones
HIGHLIGHT:      We derive a pseudo-Boolean cost function that is equivalent to a K-spin Hamiltonian representation of the discrete, finite, discounted Markov decision process with infinite horizon.

19, TITLE:      Adversarial Style Mining for One-Shot Unsupervised Domain Adaptation
http://arxiv.org/abs/2004.06042
AUTHORS:        Yawei Luo ; Ping Liu ; Tao Guan ; Junqing Yu ; Yi Yang
COMMENTS:       Preprint
HIGHLIGHT:      To this end, we propose a novel Adversarial Style Mining approach, which combines the style transfer module and task-specific module into an adversarial manner.

20, TITLE:      Adversarial Augmentation Policy Search for Domain and Cross-Lingual Generalization in Reading Comprehension
http://arxiv.org/abs/2004.06076
AUTHORS:        Adyasha Maharana ; Mohit Bansal
COMMENTS:       15 pages
HIGHLIGHT:      We address this issue via RL and more efficient Bayesian policy search methods for automatically learning the best augmentation policy combinations of the transformation probability for each adversary in a large search space.

21, TITLE:      BLEU might be Guilty but References are not Innocent
http://arxiv.org/abs/2004.06063
AUTHORS:        Markus Freitag ; David Grangier ; Isaac Caswell
HIGHLIGHT:      We study different methods to collect references and compare their value in automated evaluation by reporting correlation with human evaluation for a variety of systems and metrics.

22, TITLE:      Thinking While Moving: Deep Reinforcement Learning with Concurrent Control
http://arxiv.org/abs/2004.06089
AUTHORS:        Ted Xiao ; Eric Jang ; Dmitry Kalashnikov ; Sergey Levine ; Julian Ibarz ; Karol Hausman ; Alexander Herzog
COMMENTS:       Published as a conference paper at ICLR 2020
HIGHLIGHT:      We evaluate our methods on simulated benchmark tasks and a large-scale robotic grasping task where the robot must "think while moving".

23, TITLE:      Joint translation and unit conversion for end-to-end localization
http://arxiv.org/abs/2004.05219
AUTHORS:        Georgiana Dinu ; Prashant Mathur ; Marcello Federico ; Stanislas Lauly ; Yaser Al-Onaizan
HIGHLIGHT:      In this paper, we take unit conversions as an example and propose a data augmentation technique which leads to models learning both translation and conversion tasks as well as how to adequately switch between them for end-to-end localization.

24, TITLE:      Deep Learning for Image and Point Cloud Fusion in Autonomous Driving: A Review
http://arxiv.org/abs/2004.05224
AUTHORS:        Yaodong Cui ; Ren Chen ; Wenbo Chu ; Long Chen ; Daxin Tian ; Dongpu Cao
HIGHLIGHT:      Based on these observations, we provide our insights and point out promising research directions.

25, TITLE:      End to End Chinese Lexical Fusion Recognition with Sememe Knowledge
http://arxiv.org/abs/2004.05456
AUTHORS:        Yijiang Liu ; Meishan Zhang ; Donghong Ji
HIGHLIGHT:      In this paper, we present Chinese lexical fusion recognition, a new task which could be regarded as one kind of coreference recognition. We manually annotate a benchmark dataset for the task and then conduct experiments on it.

26, TITLE:      A Review on Deep Learning Techniques for Video Prediction
http://arxiv.org/abs/2004.05214
AUTHORS:        Sergiu Oprea ; Pablo Martinez-Gonzalez ; Alberto Garcia-Garcia ; John Alejandro Castro-Vargas ; Sergio Orts-Escolano ; Jose Garcia-Rodriguez ; Antonis Argyros
COMMENTS:       Submitted to TPAMI
HIGHLIGHT:      Motivated by the increasing interest in this task, we provide a review on the deep learning methods for prediction in video sequences.

27, TITLE:      Y-net: Biomedical Image Segmentation and Clustering
http://arxiv.org/abs/2004.05698
AUTHORS:        Sharmin Pathan ; Anant Tripathi
HIGHLIGHT:      We propose a deep clustering architecture alongside image segmentation for medical image analysis.

28, TITLE:      Bio-inspired Gait Imitation of Hexapod Robot Using Event-Based Vision Sensor and Spiking Neural Network
http://arxiv.org/abs/2004.05450
AUTHORS:        Justin Ting ; Yan Fang ; Ashwin Sanjay Lele ; Arijit Raychowdhury
COMMENTS:       7 pages, 9 figures, to be published in proceeding of IEEE WCCI/IJCNN
HIGHLIGHT:      We propose a bio-inspired feed-forward approach based on neuromorphic computing and event-based vision to address the gait imitation problem.

29, TITLE:      LAReQA: Language-agnostic answer retrieval from a multilingual pool
http://arxiv.org/abs/2004.05484
AUTHORS:        Uma Roy ; Noah Constant ; Rami Al-Rfou ; Aditya Barua ; Aaron Phillips ; Yinfei Yang
HIGHLIGHT:      We present LAReQA, a challenging new benchmark for language-agnostic answer retrieval from a multilingual candidate pool.

30, TITLE:      FLIVVER: Fly Lobula Inspired Visual Velocity Estimation & Ranging
http://arxiv.org/abs/2004.05247
AUTHORS:        Bryson Lingenfelter ; Arunava Nag ; Floris van Breugel
COMMENTS:       8 pages, 6 figures
HIGHLIGHT:      Here we present a novel algorithm, FLIVVER, which combines the geometry of dynamic forward motion with inspiration from insect visual processing to \textit{directly} estimate absolute ground velocity from a combination of optic flow and acceleration information.

31, TITLE:      Brain-inspired self-organization with cellular neuromorphic computing for multimodal unsupervised learning
http://arxiv.org/abs/2004.05488
AUTHORS:        Lyes Khacef ; Laurent Rodriguez ; Benoit Miramond
COMMENTS:       Preprint, 24 pages, 11 figures, 4 tables
HIGHLIGHT:      [...] In this paper, we build a brain-inspired neural system based on the Reentry principles, using Self-Organizing Maps and Hebbian-like learning.

32, TITLE:      Sequence Model Design for Code Completion in the Modern IDE
http://arxiv.org/abs/2004.05249
AUTHORS:        Gareth Ari Aye ; Gail E. Kaiser
HIGHLIGHT:      To meet these additional requirements, we propose a novel design for predicting top-k next tokens that combines static analysis' ability to enumerate all valid keywords and in-scope identifiers with the ability of a language model to place a probability distribution over them.

33, TITLE:      Unsupervised Commonsense Question Answering with Self-Talk
http://arxiv.org/abs/2004.05483
AUTHORS:        Vered Shwartz ; Peter West ; Ronan Le Bras ; Chandra Bhagavatula ; Yejin Choi
HIGHLIGHT:      We propose an unsupervised framework based on \emph{self-talk} as a novel alternative to multiple-choice commonsense tasks.

34, TITLE:      End-to-end Learning Improves Static Object Geo-localization in Monocular Video
http://arxiv.org/abs/2004.05232
AUTHORS:        Mohamed Chaabane ; Lionel Gueguen ; Ameni Trabelsi ; Ross Beveridge ; Stephen O'Hara
HIGHLIGHT:      In this work, we present a system that improves the localization of static objects by jointly-optimizing the components of the system via learning.

35, TITLE:      Robot self/other distinction: active inference meets neural networks learning in a mirror
http://arxiv.org/abs/2004.05473
AUTHORS:        Pablo Lanillos ; Jordi Pages ; Gordon Cheng
COMMENTS:       Accepted at European Conference on Artificial Intelligence (ECAI 2020)
HIGHLIGHT:      In this paper, we describe self-recognition as a process that is built on top of body perception unconscious mechanisms.

36, TITLE:      Classifying Constructive Comments
http://arxiv.org/abs/2004.05476
AUTHORS:        Varada Kolhatkar ; Nithum Thain ; Jeffrey Sorensen ; Lucas Dixon ; Maite Taboada
COMMENTS:       24 pages, 4 figures, 7 tables
HIGHLIGHT:      We introduce the Constructive Comments Corpus (C3), comprised of 12,000 annotated news comments, intended to help build new tools for online communities to improve the quality of their discussions.

37, TITLE:      Attend and Decode: 4D fMRI Task State Decoding Using Attention Models
http://arxiv.org/abs/2004.05234
AUTHORS:        Sam Nguyen ; Brenda Ng ; Alan K. Kaplan ; Priyadip Ray
HIGHLIGHT:      In this work, we propose to tackle the fMRI task state decoding problem by casting it as a 4D spatio-temporal classification problem.

38, TITLE:      Shape Estimation for Elongated Deformable Object using B-spline Chained Multiple Random Matrices Model
http://arxiv.org/abs/2004.05233
AUTHORS:        Gang Yao ; Ryan Saltus ; Ashwin Dani
HIGHLIGHT:      In this paper, a B-spline chained multiple random matrices representation is proposed to model geometric characteristics of an elongated deformable object.

39, TITLE:      Blind Bounded Source Separation Using Neural Networks with Local Learning Rules
http://arxiv.org/abs/2004.05479
AUTHORS:        Alper T. Erdogan ; Cengiz Pehlevan
COMMENTS:       ICASSP 2020
HIGHLIGHT:      To separate such bounded sources from their mixtures, we propose a new optimization problem, Bounded Similarity Matching (BSM).

40, TITLE:      Farmland Parcel Delineation Using Spatio-temporal Convolutional Networks
http://arxiv.org/abs/2004.05471
AUTHORS:        Han Lin Aung ; Burak Uzkent ; Marshall Burke ; David Lobell ; Stefano Ermon
HIGHLIGHT:      In this paper, we break down this task using satellite imaging into two approaches: 1) Segmentation of parcel boundaries, and 2) Segmentation of parcel areas.

41, TITLE:      What Kind of Programming Language Best Suits Integrative AGI?
http://arxiv.org/abs/2004.05267
AUTHORS:        Ben Goertzel
HIGHLIGHT:      A gradual typing approach should be used to enable mixture of rules and other metagraph nodes/links associated with various type systems, and untyped metagraph nodes/links not associated with any type system.

42, TITLE:      Grounding Occam's Razor in a Formal Theory of Simplicity
http://arxiv.org/abs/2004.05269
AUTHORS:        Ben Goertzel
HIGHLIGHT:      With this in mind, a simple formal theory of simplicity is introduced.

43, TITLE:      Combinatorial Decision Dags: A Natural Computational Model for General Intelligence
http://arxiv.org/abs/2004.05268
AUTHORS:        Ben Goertzel
HIGHLIGHT:      A novel computational model (CoDD) utilizing combinatory logic to create higher-order decision trees is presented.

44, TITLE:      Towards Anomaly Detection in Dashcam Videos
http://arxiv.org/abs/2004.05261
AUTHORS:        Sanjay Haresh ; Sateesh Kumar ; M. Zeeshan Zia ; Quoc-Huy Tran
COMMENTS:       To appear at IV 2020
HIGHLIGHT:      We propose to apply data-driven anomaly detection ideas from deep learning to dashcam videos, which hold the promise of bridging this gap. To counter this issue, we present a large and diverse dataset of truck dashcam videos, namely RetroTrucks, that includes normal and anomalous driving scenes.

45, TITLE:      Learning to Manipulate Individual Objects in an Image
http://arxiv.org/abs/2004.05495
AUTHORS:        Yanchao Yang ; Yutong Chen ; Stefano Soatto
HIGHLIGHT:      We describe a method to train a generative model with latent factors that are (approximately) independent and localized.

46, TITLE:      Pretrained Transformers Improve Out-of-Distribution Robustness
http://arxiv.org/abs/2004.06100
AUTHORS:        Dan Hendrycks ; Xiaoyuan Liu ; Eric Wallace ; Adam Dziedzic ; Rishabh Krishnan ; Dawn Song
COMMENTS:       ACL 2020
HIGHLIGHT:      We measure the generalization of previous models including bag-of-words models, ConvNets, and LSTMs, and we show that pretrained Transformers' performance declines are substantially smaller.

47, TITLE:      FDA: Fourier Domain Adaptation for Semantic Segmentation
http://arxiv.org/abs/2004.05498
AUTHORS:        Yanchao Yang ; Stefano Soatto
HIGHLIGHT:      We describe a simple method for unsupervised domain adaptation, whereby the discrepancy between the source and target distributions is reduced by swapping the low-frequency spectrum of one with the other.

48, TITLE:      Relaxed Dual Optimal Inequalities for Relaxed Columns: with Application to Vehicle Routing
http://arxiv.org/abs/2004.05499
AUTHORS:        Naveed Haghani ; Claudio Contardo ; Julian Yarkony
HIGHLIGHT:      We address the problem of accelerating column generation for set cover problems in which we relax the state space of the columns to do efficient pricing.

49, TITLE:      From Quantized DNNs to Quantizable DNNs
http://arxiv.org/abs/2004.05284
AUTHORS:        Kunyuan Du ; Ya Zhang ; Haibing Guan
HIGHLIGHT:      This paper proposes Quantizable DNNs, a special type of DNNs that can flexibly quantize its bit-width (denoted as `bit modes' thereafter) during execution without further re-training.

50, TITLE:      Multi-View Matching (MVM): Facilitating Multi-Person 3D Pose Estimation Learning with Action-Frozen People Video
http://arxiv.org/abs/2004.05275
AUTHORS:        Yeji Shen ; C. -C. Jay Kuo
COMMENTS:       16 pages, 6 figures, submitted JVCI
HIGHLIGHT:      To tackle the challeging problem of multi-person 3D pose estimation from a single image, we propose a multi-view matching (MVM) method in this work.

51, TITLE:      Improved Speech Representations with Multi-Target Autoregressive Predictive Coding
http://arxiv.org/abs/2004.05274
AUTHORS:        Yu-An Chung ; James Glass
COMMENTS:       Accepted to ACL 2020
HIGHLIGHT:      In this paper we extend this hypothesis and aim to enrich the information encoded in the hidden states by training the model to make more accurate future predictions.

52, TITLE:      Regularizing Meta-Learning via Gradient Dropout
http://arxiv.org/abs/2004.05859
AUTHORS:        Hung-Yu Tseng ; Yi-Wen Chen ; Yi-Hsuan Tsai ; Sifei Liu ; Yen-Yu Lin ; Ming-Hsuan Yang
COMMENTS:       Code: https://github.com/hytseng0509/DropGrad
HIGHLIGHT:      In this paper, we introduce a simple yet effective method to alleviate the risk of overfitting for gradient-based meta-learning.

53, TITLE:      $\texttt{ArCOV-19}$: The First Arabic COVID-19 Twitter Dataset with Propagation Networks
http://arxiv.org/abs/2004.05861
AUTHORS:        Fatima Haouari ; Maram Hasanain ; Reem Suwaileh ; Tamer Elsayed
HIGHLIGHT:      In this paper, we present $\texttt{ArCOV-19}$, an Arabic COVID-19 Twitter dataset that covers the period from 27$^{th}$ of January till 31$^{st}$ of March 2020.

54, TITLE:      SSP: Single Shot Future Trajectory Prediction
http://arxiv.org/abs/2004.05846
AUTHORS:        Isht Dwivedi ; Srikanth Malla ; Behzad Dariush ; Chiho Choi
HIGHLIGHT:      We propose a robust solution to future trajectory forecast, which can be practically applicable to autonomous agents in highly crowded environments.

55, TITLE:      Relational Learning between Multiple Pulmonary Nodules via Deep Set Attention Transformers
http://arxiv.org/abs/2004.05640
AUTHORS:        Jiancheng Yang ; Haoran Deng ; Xiaoyang Huang ; Bingbing Ni ; Yi Xu
COMMENTS:       2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI 2020)
HIGHLIGHT:      In this study, we propose a multiple instance learning (MIL) approach and empirically prove the benefit to learn the relations between multiple nodules.

56, TITLE:      Revisiting Loss Landscape for Adversarial Robustness
http://arxiv.org/abs/2004.05884
AUTHORS:        Dongxian Wu ; Yisen Wang ; Shu-tao Xia
HIGHLIGHT:      In this paper, we investigate the surface geometry of several well-recognized adversarial training variants, and reveal that their adversarial loss landscape is closely related to the adversarially robust generalization, i.e., the flatter the adversarial loss landscape, the smaller the adversarially robust generalization gap.

57, TITLE:      Frequency-Guided Word Substitutions for Detecting Textual Adversarial Examples
http://arxiv.org/abs/2004.05887
AUTHORS:        Maximilian Mozes ; Pontus Stenetorp ; Bennett Kleinberg ; Lewis D. Griffin
COMMENTS:       pre-print
HIGHLIGHT:      Based on these findings, we propose frequency-guided word substitutions (FGWS) as a simple algorithm for the automatic detection of adversarially perturbed textual sequences.

58, TITLE:      Residual Attention U-Net for Automated Multi-Class Segmentation of COVID-19 Chest CT Images
http://arxiv.org/abs/2004.05645
AUTHORS:        Xiaocong Chen ; Lina Yao ; Yu Zhang
HIGHLIGHT:      To this end, we proposed a novel deep learning algorithm for automated segmentation of multiple COVID-19 infection regions.

59, TITLE:      Analysis of The Ratio of $\ell_1$ and $\ell_2$ Norms in Compressed Sensing
http://arxiv.org/abs/2004.05873
AUTHORS:        Yiming Xu ; Akil Narayan ; Hoang Tran ; Clayton Webster
COMMENTS:       24 pages, 5 figures
HIGHLIGHT:      We call this initialization approach \emph{support selection}, and we demonstrate that it empirically improves the performance of existing $\ell_1/\ell_2$ algorithms.

60, TITLE:      Optimizing Reachability Sets in Temporal Graphs by Delaying
http://arxiv.org/abs/2004.05875
AUTHORS:        Argyrios Deligkas ; Igor Potapov
HIGHLIGHT:      In this paper, we study how changes of the time labels, corresponding to delays on the availability of the edges, affect the reachability sets from given sources.

61, TITLE:      Optimal Learning for Sequential Decisions in Laboratory Experimentation
http://arxiv.org/abs/2004.05417
AUTHORS:        Kris Reyes ; Warren B Powell
HIGHLIGHT:      We introduce the concept of a learning policy, and review the major categories of policies.

62, TITLE:      The Role of Stem Noise in Visual Perception and Image Quality Measurement
http://arxiv.org/abs/2004.05422
AUTHORS:        Arash Ashtari
COMMENTS:       16 pages,19 figures
HIGHLIGHT:      This paper considers reference free quality assessment of distorted and noisy images.

63, TITLE:      Unveiling COVID-19 from Chest X-ray with deep learning: a hurdles race with small data
http://arxiv.org/abs/2004.05405
AUTHORS:        Enzo Tartaglione ; Carlo Alberto Barbano ; Claudio Berzovini ; Marco Calandri ; Marco Grangetto
HIGHLIGHT:      In this study we provide insights and also raise warnings on what is reasonable to expect by applying deep-learning to COVID classification of CXR images.

64, TITLE:      Workflow Automation for Cyber Physical System Development Processes
http://arxiv.org/abs/2004.05654
AUTHORS:        Charles Hartsell ; Nagabhushan Mahadevan ; Harmon Nine ; Ted Bapty ; Abhishek Dubey ; Gabor Karsai
COMMENTS:       Accepted for Publication at DESTION 2020
HIGHLIGHT:      To address this problem, we introduce a workflow modeling language for the automation of complex CPS development processes and implement a platform for execution of these models in the Assurance-based Learning-enabled CPS (ALC) Toolchain.

65, TITLE:      Annotating Social Determinants of Health Using Active Learning, and Characterizing Determinants Using Neural Event Extraction
http://arxiv.org/abs/2004.05438
AUTHORS:        Kevin Lybarger ; Mari Ostendorf ; Meliha Yetisgen
COMMENTS:       29 pages, 14 figures, 4 tables
HIGHLIGHT:      This work presents a new corpus with SDOH annotations, a novel active learning framework, and the first extraction results on the new corpus.

66, TITLE:      MLCVNet: Multi-Level Context VoteNet for 3D Object Detection
http://arxiv.org/abs/2004.05679
AUTHORS:        Qian Xie ; Yu-Kun Lai ; Jing Wu ; Zhoutao Wang ; Yiming Zhang ; Kai Xu ; Jun Wang
COMMENTS:       To be presented at CVPR 2020
HIGHLIGHT:      In this paper, we address the 3D object detection task by capturing multi-level contextual information with the self-attention mechanism and multi-scale feature fusion.

67, TITLE:      PatchAttack: A Black-box Texture-based Attack with Reinforcement Learning
http://arxiv.org/abs/2004.05682
AUTHORS:        Chenglin Yang ; Adam Kortylewski ; Cihang Xie ; Yinzhi Cao ; Alan Yuille
HIGHLIGHT:      PatchAttack: A Black-box Texture-based Attack with Reinforcement Learning

68, TITLE:      Low-Resolution Overhead Thermal Tripwire for Occupancy Estimation
http://arxiv.org/abs/2004.05685
AUTHORS:        Mertcan Cokbas ; Prakash Ishwar ; Janusz Konrad
HIGHLIGHT:      We propose a people counting system which uses a low-resolution thermal sensor.

69, TITLE:      TinyMBERT: Multi-Stage Distillation Framework for Massive Multi-lingual NER
http://arxiv.org/abs/2004.05686
AUTHORS:        Subhabrata Mukherjee ; Ahmed Awadallah
COMMENTS:       To appear in ACL 2020
HIGHLIGHT:      In this work we study knowledge distillation with a focus on multi-lingual Named Entity Recognition (NER).

70, TITLE:      Implicit Multi-Agent Coordination at Unsignalized Intersections via Topological Inference
http://arxiv.org/abs/2004.05205
AUTHORS:        Christoforos Mavrogiannis ; Jonathan A. DeCastro ; Siddhartha S. Srinivasa
HIGHLIGHT:      In this paper, we do so by representing modes of joint behavior as topological braids.

71, TITLE:      Underwater Image Enhancement Based on Structure-Texture Reconstruction
http://arxiv.org/abs/2004.05430
AUTHORS:        Sen Lin ; Kaichen Chi
HIGHLIGHT:      Underwater Image Enhancement Based on Structure-Texture Reconstruction

72, TITLE:      Detection of Covid-19 From Chest X-ray Images Using Artificial Intelligence: An Early Review
http://arxiv.org/abs/2004.05436
AUTHORS:        Muhammad Ilyas ; Hina Rehman ; Amine Nait-ali
HIGHLIGHT:      In this article we will discuss the different approaches used for the detection of COVID-19 and the challenges we are facing.

73, TITLE:      Unsupervised Facial Action Unit Intensity Estimation via Differentiable Optimization
http://arxiv.org/abs/2004.05908
AUTHORS:        Xinhui Song ; Tianyang Shi ; Tianjia Shao ; Yi Yuan ; Zunlei Feng ; Changjie Fan
HIGHLIGHT:      Considering all these difficulties, we propose an unsupervised framework GE-Net for facial AU intensity estimation from a single image, without requiring any annotated AU data.

74, TITLE:      k-decay: A New Method For Learning Rate Schedule
http://arxiv.org/abs/2004.05909
AUTHORS:        Tao Zhang
HIGHLIGHT:      In the paper, evaluate the k-decay method by the new polynomial function on CIFAR-10 and CIFAR-100 datasets with different neural networks (ResNet, Wide ResNet and DenseNet), the results improvements over the state-of-the-art results on most of them.

75, TITLE:      Knowledge Distillation and Student-Teacher Learning for Visual Intelligence: A Review and New Outlooks
http://arxiv.org/abs/2004.05937
AUTHORS:        Lin Wang ; Kuk-Jin Yoon
COMMENTS:       30 pages, paper in submission
HIGHLIGHT:      In general, we consider some fundamental questions that have been driving this research area and thoroughly generalize the research progress and technical details.

76, TITLE:      Technical Report: NEMO DNN Quantization for Deployment Model
http://arxiv.org/abs/2004.05930
AUTHORS:        Francesco Conti
COMMENTS:       12 pages, technical report
HIGHLIGHT:      This technical report aims at defining a formal framework for Deep Neural Network (DNN) layer-wise quantization, focusing in particular on the problems related to the final deployment.

77, TITLE:      Learning Spatial Relationships between Samples of Image Shapes
http://arxiv.org/abs/2004.05713
AUTHORS:        Juan Castorena ; Manish Bhattarai ; Diane Oyen
HIGHLIGHT:      Here, we propose a method that combines sparsely sampling points from image shapes and learning the spatial relationships between the extracted samples that characterize them.

78, TITLE:      Towards an Efficient Deep Learning Model for COVID-19 Patterns Detection in X-ray Images
http://arxiv.org/abs/2004.05717
AUTHORS:        Eduardo Luz ; Pedro Lopes Silva ; Rodrigo Silva ; Gladston Moreira
COMMENTS:       Copyright 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
HIGHLIGHT:      Thus, in this work, we address the hypothesis that better performance in terms of overall accuracy and COVID-19 sensitivity can be achieved with much more compact models.

79, TITLE:      Principal Neighbourhood Aggregation for Graph Nets
http://arxiv.org/abs/2004.05718
AUTHORS:        Gabriele Corso ; Luca Cavalleri ; Dominique Beaini ; Pietro Liò ; Petar Veličković
HIGHLIGHT:      Accordingly, we propose Principal Neighbourhood Aggregation (PNA), a novel architecture combining multiple aggregators with degree-scalers (which generalize the sum aggregator).

80, TITLE:      Keyword Assisted Topic Models
http://arxiv.org/abs/2004.05964
AUTHORS:        Shusei Eshima ; Kosuke Imai ; Tomoya Sasaki
HIGHLIGHT:      In this paper, we empirically demonstrate that providing topic models with a small number of keywords can substantially improve their performance.

81, TITLE:      Complaint-driven Training Data Debugging for Query 2.0
http://arxiv.org/abs/2004.05722
AUTHORS:        Weiyuan Wu ; Lampros Flokas ; Eugene Wu ; Jiannan Wang
COMMENTS:       Proceedings of the 2020 ACM SIGMOD International Conference on Management of Data
HIGHLIGHT:      We propose two novel heuristic approaches based on influence functions which both require linear retraining steps.

82, TITLE:      A negative case analysis of visual grounding methods for VQA
http://arxiv.org/abs/2004.05704
AUTHORS:        Robik Shrestha ; Kushal Kafle ; Christopher Kanan
HIGHLIGHT:      To address this issue, recent bias mitigation methods for VQA propose to incorporate visual cues (e.g., human attention maps) to better ground the VQA models, showcasing impressive gains.

83, TITLE:      From Holant to Quantum Entanglement and Back
http://arxiv.org/abs/2004.05706
AUTHORS:        Jin-Yi Cai ; Zhiguo Fu ; Shuai Shao
COMMENTS:       46 pages
HIGHLIGHT:      We discover two particular entangled states $|{\Psi_6}\rangle$ of 6 qubits and $|{\Psi_8}\rangle$ of 8 qubits respectively, that have extraordinary and unique closure properties in terms of the Bell property.

84, TITLE:      VGCN-BERT: Augmenting BERT with Graph Embedding for Text Classification
http://arxiv.org/abs/2004.05707
AUTHORS:        Zhibin Lu ; Pan Du ; Jian-Yun Nie
COMMENTS:       12 pages, 2 figures
HIGHLIGHT:      In this paper, we propose VGCN-BERT model which combines the capability of BERT with a Vocabulary Graph Convolutional Network (VGCN).

85, TITLE:      Exploit Where Optimizer Explores via Residuals
http://arxiv.org/abs/2004.05298
AUTHORS:        An Xu ; Zhouyuan Huo ; Heng Huang
HIGHLIGHT:      In this work we propose a novel optimization method named (momentum) stochastic gradient descent with residuals (RSGD(m)) to exploit the gradient descent trajectory using proper residual schemes, which leads to a performance boost of both the convergence and generalization.

86, TITLE:      Spatially-Attentive Patch-Hierarchical Network for Adaptive Motion Deblurring
http://arxiv.org/abs/2004.05343
AUTHORS:        Maitreya Suin ; Kuldeep Purohit ; A. N. Rajagopalan
COMMENTS:       Accepted at CVPR2020
HIGHLIGHT:      In this work, we propose an efficient pixel adaptive and feature attentive design for handling large blur variations across different spatial locations and process each test image adaptively.

87, TITLE:      Sharing Matters for Generalization in Deep Metric Learning
http://arxiv.org/abs/2004.05582
AUTHORS:        Timo Milbich ; Karsten Roth ; Biagio Brattoli ; Björn Ommer
COMMENTS:       Technical Report
HIGHLIGHT:      This work investigates how to learn such characteristics without the need for extra annotations or training data.

88, TITLE:      DeepSentiPers: Novel Deep Learning Models Trained Over Proposed Augmented Persian Sentiment Corpus
http://arxiv.org/abs/2004.05328
AUTHORS:        Javad PourMostafa Roshan Sharami ; Parsa Abbasi Sarabestani ; Seyed Abolghasem Mirroshandel
HIGHLIGHT:      Our comprehensive experiments on three baselines and two different neural word embedding methods show that our data augmentation methods and intended models successfully address the aims of the research.

89, TITLE:      Explaining Question Answering Models through Text Generation
http://arxiv.org/abs/2004.05569
AUTHORS:        Veronica Latcinnik ; Jonathan Berant
HIGHLIGHT:      In this work, we propose a model for multi-choice question answering, where a LM-based generator generates a textual hypothesis that is later used by a classifier to answer the question.

90, TITLE:      YouMakeup VQA Challenge: Towards Fine-grained Action Understanding in Domain-Specific Videos
http://arxiv.org/abs/2004.05573
AUTHORS:        Shizhe Chen ; Weiying Wang ; Ludan Ruan ; Linli Yao ; Qin Jin
COMMENTS:       CVPR LVVU Workshop 2020 YouMakeup VQA Challenge
HIGHLIGHT:      In this paper, we present the challenge guidelines, the dataset used, and performances of baseline models on the two proposed tasks.

91, TITLE:      AMR Parsing via Graph-Sequence Iterative Inference
http://arxiv.org/abs/2004.05572
AUTHORS:        Deng Cai ; Wai Lam
COMMENTS:       ACL2020
HIGHLIGHT:      We propose a new end-to-end model that treats AMR parsing as a series of dual decisions on the input sequence and the incrementally constructed graph.

92, TITLE:      Image Co-skeletonization via Co-segmentation
http://arxiv.org/abs/2004.05575
AUTHORS:        Koteswar Rao Jerripothula ; Jianfei Cai ; Jiangbo Lu ; Junsong Yuan
COMMENTS:       13 pages, 12 figures, Submitted to IEEE Transactions on Image Processing (TIP)
HIGHLIGHT:      Therefore, we propose a coupled framework for co-skeletonization and co-segmentation tasks so that they are well informed by each other, and benefit each other synergistically. Since it is a new problem, we also construct a benchmark dataset by annotating nearly 1.8k images spread across 38 categories.

93, TITLE:      When Weak Becomes Strong: Robust Quantification of White Matter Hyperintensities in Brain MRI scans
http://arxiv.org/abs/2004.05578
AUTHORS:        Oliver Werner ; Kimberlin M. H. van Wijnen ; Wiro J. Niessen ; Marius de Groot ; Meike W. Vernooij ; Florian Dubost ; Marleen de Bruijne
COMMENTS:       11 pages, 3 figures
HIGHLIGHT:      In this article, we compared networks optimized with weak and strong labels, and study their ability to generalize to other datasets.

94, TITLE:      Cross-domain Correspondence Learning for Exemplar-based Image Translation
http://arxiv.org/abs/2004.05571
AUTHORS:        Pan Zhang ; Bo Zhang ; Dong Chen ; Lu Yuan ; Fang Wen
COMMENTS:       Accepted as a CVPR 2020 oral paper
HIGHLIGHT:      We present a general framework for exemplar-based image translation, which synthesizes a photo-realistic image from the input in a distinct domain (e.g., semantic segmentation mask, or edge map, or pose keypoints), given an exemplar image.

95, TITLE:      Which visual questions are difficult to answer? Analysis with Entropy of Answer Distributions
http://arxiv.org/abs/2004.05595
AUTHORS:        Kento Terao ; Toru Tamaki ; Bisser Raytchev ; Kazufumi Kaneda ; Shun'ichi Satoh
HIGHLIGHT:      We propose a novel approach to identify the difficulty of visual questions for Visual Question Answering (VQA) without direct supervision or annotations to the difficulty.

96, TITLE:      Exploring The Spatial Reasoning Ability of Neural Models in Human IQ Tests
http://arxiv.org/abs/2004.05352
AUTHORS:        Hyunjae Kim ; Yookyung Koh ; Jinheon Baek ; Jaewoo Kang
HIGHLIGHT:      In this work, we focus on spatial reasoning and explore the spatial understanding of neural models. Using well-defined rules, we constructed datasets that consist of various complexity levels.

97, TITLE:      You Impress Me: Dialogue Generation via Mutual Persona Perception
http://arxiv.org/abs/2004.05388
AUTHORS:        Qian Liu ; Yihong Chen ; Bei Chen ; Jian-Guang Lou ; Zixuan Chen ; Bin Zhou ; Dongmei Zhang
COMMENTS:       Accepted by ACL 2020, code is avaiable at https://github.com/SivilTaram/Persona-Dialogue-Generation
HIGHLIGHT:      Motivated by this, we propose P^2 Bot, a transmitter-receiver based framework with the aim of explicitly modeling understanding.

98, TITLE:      Training Data Set Assessment for Decision-Making in a Multiagent Landmine Detection Platform
http://arxiv.org/abs/2004.05380
AUTHORS:        Johana Florez-Lozano ; Fabio Caraffini ; Carlos Parra ; Mario Gongora
HIGHLIGHT:      A novel approach to solve these problems includes distributed systems, as presented in this work based on hardware and software multi-agent systems.

99, TITLE:      Bayesian Surprise in Indoor Environments
http://arxiv.org/abs/2004.05381
AUTHORS:        Sebastian Feld ; Andreas Sedlmeier ; Markus Friedrich ; Jan Franz ; Lenz Belzner
COMMENTS:       10 pages, 16 figures
HIGHLIGHT:      This paper proposes a novel method to identify unexpected structures in 2D floor plans using the concept of Bayesian Surprise.

100, TITLE:     Trajectory annotation using sequences of spatial perception
http://arxiv.org/abs/2004.05383
AUTHORS:        Sebastian Feld ; Steffen Illium ; Andreas Sedlmeier ; Lenz Belzner
COMMENTS:       10 pages, 17 figures
HIGHLIGHT:      We propose an unsupervised learning approach based on a neural autoencoding that learns semantically meaningful continuous encodings of spatio-temporal trajectory data.

101, TITLE:     Toward Subgraph Guided Knowledge Graph Question Generation with Graph Neural Networks
http://arxiv.org/abs/2004.06015
AUTHORS:        Yu Chen ; Lingfei Wu ; Mohammed J. Zaki
COMMENTS:       12 pages
HIGHLIGHT:      In this work, we focus on a more realistic setting, where we aim to generate questions from a KG subgraph and target answers.

102, TITLE:     Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training
http://arxiv.org/abs/2004.06002
AUTHORS:        Hongkai Zhang ; Hong Chang ; Bingpeng Ma ; Naiyan Wang ; Xilin Chen
HIGHLIGHT:      In this work, we first point out the inconsistency problem between the fixed network settings and the dynamic training procedure, which greatly affects the performance.

103, TITLE:     Analysing Flow Security Properties in Virtualised Computing Systems
http://arxiv.org/abs/2004.05500
AUTHORS:        Chunyan Mu
HIGHLIGHT:      We propose a distributed process algebra CSP_{4v} with security labelled processes for the purpose of formal modelling of virtualised computing systems.

104, TITLE:     CLUE: A Chinese Language Understanding Evaluation Benchmark
http://arxiv.org/abs/2004.05986
AUTHORS:        Liang Xu ; Xuanwei Zhang ; Lu Li ; Hai Hu ; Chenjie Cao ; Weitang Liu ; Junyi Li ; Yudong Li ; Kai Sun ; Yechen Xu ; Yiming Cui ; Cong Yu ; Qianqian Dong ; Yin Tian ; Dian Yu ; Bo Shi ; Jun Zeng ; Rongzhao Wang ; Weijian Xie ; Yanting Li ; Yina Patterson ; Zuoyu Tian ; Yiwen Zhang ; He Zhou ; Shaoweihua Liu ; Qipeng Zhao ; Cong Yue ; Xinrui Zhang ; Zhengliang Yang ; Zhenzhong Lan
COMMENTS:       9 pages, 4 figures
HIGHLIGHT:      We introduce CLUE, a Chinese Language Understanding Evaluation benchmark. We release CLUE, baselines and pre-training dataset on Github.

105, TITLE:     Integrated Eojeol Embedding for Erroneous Sentence Classification in Korean Chatbots
http://arxiv.org/abs/2004.05744
AUTHORS:        DongHyun Choi ; IlNam Park ; Myeong Cheol Shin ; EungGyun Kim ; Dong Ryeol Shin
COMMENTS:       9 pages, 2 figures
HIGHLIGHT:      This paper proposes a novel approach of Integrated Eojeol (Korean syntactic word separated by space) Embedding to reduce the effect that poorly analyzed morphemes may make on sentence classification.

106, TITLE:     Punctuation Prediction in Spontaneous Conversations: Can We Mitigate ASR Errors with Retrofitted Word Embeddings?
http://arxiv.org/abs/2004.05985
AUTHORS:        Łukasz Augustyniak ; Piotr Szymanski ; Mikołaj Morzy ; Piotr Zelasko ; Adrian Szymczak ; Jan Mizgajski ; Yishay Carmiel ; Najim Dehak
COMMENTS:       submitted to INTERSPEECH'20
HIGHLIGHT:      Our main contribution is a method for better alignment of homonym embeddings and the validation of the presented method on the punctuation prediction task.

107, TITLE:     Speak2Label: Using Domain Knowledge for Creating a Large Scale Driver Gaze Zone Estimation Dataset
http://arxiv.org/abs/2004.05973
AUTHORS:        Shreya Ghosh ; Abhinav Dhall ; Garima Sharma ; Sarthak Gupta ; Nicu Sebe
HIGHLIGHT:      In this paper, a fully automatic technique for labelling an image based gaze behavior dataset for driver gaze zone estimation is proposed.

108, TITLE:     A Comparison of Deep Learning Convolution Neural Networks for Liver Segmentation in Radial Turbo Spin Echo Images
http://arxiv.org/abs/2004.05731
AUTHORS:        Lavanya Umapathy ; Mahesh Bharath Keerthivasan ; Jean-Phillipe Galons ; Wyatt Unger ; Diego Martin ; Maria I Altbach ; Ali Bilgin
COMMENTS:       3 pages, 4 figures, 1 table. Published in Proceedings of International Society for Magnetic Resonance in Medicine 2018
HIGHLIGHT:      In this work, we use a deep-learning convolutional neural network (CNN) for the segmentation of liver in abdominal RADTSE images.

109, TITLE:     Dense Registration and Mosaicking of Fingerprints by Training an End-to-End Network
http://arxiv.org/abs/2004.05972
AUTHORS:        Zhe Cui ; Jianjiang Feng ; Jie Zhou
HIGHLIGHT:      To overcome the limitation of handcraft features, we propose to train an end-to-end network to directly output pixel-wise displacement field between two fingerprints.

110, TITLE:     Exponential Upper Bounds for the Runtime of Randomized Search Heuristics
http://arxiv.org/abs/2004.05733
AUTHORS:        Benjamin Doerr
HIGHLIGHT:      We argue that proven exponential upper bounds on runtimes, an established area in classic algorithms, are interesting also in evolutionary computation and we prove several such results.

111, TITLE:     Reinforced Curriculum Learning on Pre-trained Neural Machine Translation Models
http://arxiv.org/abs/2004.05757
AUTHORS:        Mingjun Zhao ; Haijiang Wu ; Di Niu ; Xiaoli Wang
COMMENTS:       Accepted as full paper by AAAI-2020 (oral presentation)
HIGHLIGHT:      In this paper, we aim to learn a curriculum for improving a pre-trained NMT model by re-selecting influential data samples from the original training set and formulate this task as a reinforcement learning problem.

112, TITLE:     When Does Unsupervised Machine Translation Work?
http://arxiv.org/abs/2004.05516
AUTHORS:        Kelly Marchisio ; Kevin Duh ; Philipp Koehn
HIGHLIGHT:      We find that performance rapidly deteriorates when source and target corpora are from different domains, and that random word embedding initialization can dramatically affect downstream translation performance.

113, TITLE:     Deep Learning COVID-19 Features on CXR using Limited Training Data Sets
http://arxiv.org/abs/2004.05758
AUTHORS:        Yujin Oh ; Sangjoon Park ; Jong Chul Ye
HIGHLIGHT:      To address this problem, here we propose a patch-based convolutional neural network approach with a relatively small number of trainable parameters for COVID-19 diagnosis.

114, TITLE:     Density Map Guided Object Detection in Aerial Images
http://arxiv.org/abs/2004.05520
AUTHORS:        Changlin Li ; Taojiannan Yang ; Sijie Zhu ; Chen Chen ; Shanyue Guan
COMMENTS:       CVPR 2020 EarthVision Workshop
HIGHLIGHT:      In this paper, we investigate the image cropping strategy to address these challenges.

115, TITLE:     Augmentation of the Reconstruction Performance of Fuzzy C-Means with an Optimized Fuzzification Factor Vector
http://arxiv.org/abs/2004.05764
AUTHORS:        Kaijie Xu ; Witold Pedrycz ; Zhiwu Li
HIGHLIGHT:      In this paper, to enhance the quality of the degranulation (reconstruction) process, we augment the FCM-based degranulation mechanism by introducing a vector of fuzzification factors (fuzzification factor vector) and setting up an adjustment mechanism to modify the prototypes and the partition matrix.

116, TITLE:     UC-Net: Uncertainty Inspired RGB-D Saliency Detection via Conditional Variational Autoencoders
http://arxiv.org/abs/2004.05763
AUTHORS:        Jing Zhang ; Deng-Ping Fan ; Yuchao Dai ; Saeed Anwar ; Fatemeh Sadat Saleh ; Tong Zhang ; Nick Barnes
COMMENTS:       Accepted by IEEE CVPR 2020 (ORAL). Code: https://github.com/JingZhang617/UCNet
HIGHLIGHT:      In this paper, we propose the first framework (UCNet) to employ uncertainty for RGB-D saliency detection by learning from the data labeling process.

117, TITLE:     DeepEDN: A Deep Learning-based Image Encryption and Decryption Network for Internet of Medical Things
http://arxiv.org/abs/2004.05523
AUTHORS:        Yi Ding ; Guozheng Wu ; Dajiang Chen ; Ning Zhang ; Linpeng Gong ; Mingsheng Cao ; Zhiguang Qin
HIGHLIGHT:      In this work, a deep learning based encryption and decryption network (DeepEDN) is proposed to fulfill the process of encrypting and decrypting the medical image.

118, TITLE:     Enabling Incremental Knowledge Transfer for Object Detection at the Edge
http://arxiv.org/abs/2004.05746
AUTHORS:        Mohammad Farhadi Bajestani ; Mehdi Ghasemi ; Sarma Vrudhula ; Yezhou Yang
COMMENTS:       2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshop (CVPRW)
HIGHLIGHT:      In this paper, a system-level design is proposed to improve the energy consumption of object detection on the user-end device.

119, TITLE:     Deep Siamese Domain Adaptation Convolutional Neural Network for Cross-domain Change Detection in Multispectral Images
http://arxiv.org/abs/2004.05745
AUTHORS:        Hongruixuan Chen ; Chen Wu ; Bo Du ; Liangepei Zhang
HIGHLIGHT:      In this paper, we propose a novel deep siamese domain adaptation convolutional neural network (DSDANet) architecture for cross-domain change detection.

120, TITLE:     Data augmentation using generative networks to identify dementia
http://arxiv.org/abs/2004.05989
AUTHORS:        Bahman Mirheidari ; Yilin Pan ; Daniel Blackburn ; Ronan O'Malley ; Traci Walker ; Annalena Venneri ; Markus Reuber ; Heidi Christensen
HIGHLIGHT:      In this paper, we investigate the application of a similar approach to different types of speech and audio-based features extracted from interactions recorded with our automatic dementia detection system.

121, TITLE:     MetaIQA: Deep Meta-learning for No-Reference Image Quality Assessment
http://arxiv.org/abs/2004.05508
AUTHORS:        Hancheng Zhu ; Leida Li ; Jinjian Wu ; Weisheng Dong ; Guangming Shi
HIGHLIGHT:      With this motivation, this paper presents a no-reference IQA metric based on deep meta-learning.

122, TITLE:     Self-supervised Feature Learning by Cross-modality and Cross-view Correspondences
http://arxiv.org/abs/2004.05749
AUTHORS:        Longlong Jing ; Yucheng Chen ; Ling Zhang ; Mingyi He ; Yingli Tian
HIGHLIGHT:      Unlike most existing self-supervised methods to learn only 2D image features or only 3D point cloud features, this paper presents a novel and effective self-supervised learning approach to jointly learn both 2D image features and 3D point cloud features by exploiting cross-modality and cross-view correspondences without using any human annotated labels.

123, TITLE:     A Novel Pose Proposal Network and Refinement Pipeline for Better Object Pose Estimation
http://arxiv.org/abs/2004.05507
AUTHORS:        Ameni Trabelsi ; Mohamed Chaabane ; Nathaniel Blanchard ; Ross Beveridge
HIGHLIGHT:      In this paper, we present a novel deep learning pipeline for 6D object pose estimation and refinement from RGB inputs.

124, TITLE:     A Survey of Single-Scene Video Anomaly Detection
http://arxiv.org/abs/2004.05993
AUTHORS:        Bharathkumar Ramachandra ; Michael J. Jones ; Ranga Raju Vatsavai
HIGHLIGHT:      This survey article summarizes research trends on the topic of anomaly detection in video feeds of a single scene.

125, TITLE:     Verification of Deep Convolutional Neural Networks Using ImageStars
http://arxiv.org/abs/2004.05511
AUTHORS:        Hoang-Dung Tran ; Stanley Bak ; Weiming Xiang ; Taylor T. Johnson
HIGHLIGHT:      In this paper, we describe a set-based framework that successfully deals with real-world CNNs, such as VGG16 and VGG19, that have high accuracy on ImageNet.

126, TITLE:     Aspect and Opinion Aware Abstractive Review Summarization with Reinforced Hard Typed Decoder
http://arxiv.org/abs/2004.05755
AUTHORS:        Yufei Tian ; Jianfei Yu ; Jing Jiang
HIGHLIGHT:      In this paper, we study abstractive review summarization.Observing that review summaries often consist of aspect words, opinion words and context words, we propose a two-stage reinforcement learning approach, which first predicts the output word type from the three types, and then leverages the predicted word type to generate the final word distribution.Experimental results on two Amazon product review datasets demonstrate that our method can consistently outperform several strong baseline approaches based on ROUGE scores.

127, TITLE:     Reinforcement Learning via Reasoning from Demonstration
http://arxiv.org/abs/2004.05512
AUTHORS:        Lisa Torrey
COMMENTS:       Adaptive and Learning Agents Workshop 2020
HIGHLIGHT:      This paper proposes a framework for agents that benefit from demonstration in this human-inspired way.

128, TITLE:     Self-Supervised Tuning for Few-Shot Segmentation
http://arxiv.org/abs/2004.05538
AUTHORS:        Kai Zhu ; Wei Zhai ; Zheng-Jun Zha ; Yang Cao
COMMENTS:       under review
HIGHLIGHT:      To address this issue, this paper presents an adaptive tuning framework, in which the distribution of latent features across different episodes is dynamically adjusted based on a self-segmentation scheme, augmenting category-specific descriptors for label prediction.

129, TITLE:     Individual Tooth Detection and Identification from Dental Panoramic X-Ray Images via Point-wise Localization and Distance Regularization
http://arxiv.org/abs/2004.05543
AUTHORS:        Minyoung Chung ; Jusang Lee ; Sanguk Park ; Minkyung Lee ; Chae Eun Lee ; Jeongjin Lee ; Yeong-Gil Shin
COMMENTS:       10 pages, 7 figures
HIGHLIGHT:      In this study, we propose a point-wise tooth localization neural network by introducing a spatial distance regularization loss.

130, TITLE:     Inter-Region Affinity Distillation for Road Marking Segmentation
http://arxiv.org/abs/2004.05304
AUTHORS:        Yuenan Hou ; Zheng Ma ; Chunxiao Liu ; Tak-Wai Hui ; Chen Change Loy
COMMENTS:       10 pages, 10 figures; This paper is accepted by CVPR 2020; Our code is available at https://github.com/cardwing/Codes-for-IntRA-KD
HIGHLIGHT:      In this work, we explore a novel knowledge distillation (KD) approach that can transfer 'knowledge' on scene structure more effectively from a teacher to a student model.

131, TITLE:     Object-oriented SLAM using Quadrics and Symmetry Properties for Indoor Environments
http://arxiv.org/abs/2004.05303
AUTHORS:        Ziwei Liao ; Wei Wang ; Xianyu Qi ; Xiaoyu Zhang ; Lin Xue ; Jianzhen Jiao ; Ran Wei
COMMENTS:       Submission to IROS 2020. Video: https://youtu.be/u9zRBp4TPIs
HIGHLIGHT:      Aiming at the application environment of indoor mobile robots, this paper proposes a sparse object-level SLAM algorithm based on an RGB-D camera.

132, TITLE:     Building Disaster Damage Assessment in Satellite Imagery with Multi-Temporal Fusion
http://arxiv.org/abs/2004.05525
AUTHORS:        Ethan Weber ; Hassan Kané
COMMENTS:       Accepted for presentation at the ICLR 2020 AI For Earth Sciences Workshop
HIGHLIGHT:      In this work, we report findings on problem framing, data processing and training procedures which are specifically helpful for the task of building damage assessment using the newly released xBD dataset.

133, TITLE:     Gradients as Features for Deep Representation Learning
http://arxiv.org/abs/2004.05529
AUTHORS:        Fangzhou Mu ; Yingyu Liang ; Yin Li
COMMENTS:       ICLR 2020 conference paper
HIGHLIGHT:      Specifically, we propose to explore gradient-based features.

134, TITLE:     Generating Fact Checking Explanations
http://arxiv.org/abs/2004.05773
AUTHORS:        Pepa Atanasova ; Jakob Grue Simonsen ; Christina Lioma ; Isabelle Augenstein
COMMENTS:       In Proceedings of the 2020 Annual Conference of the Association for Computational Linguistics (ACL 2020)
HIGHLIGHT:      This paper provides the first study of how these explanations can be generated automatically based on available claim context, and how this task can be modelled jointly with veracity prediction.

135, TITLE:     A Unified DNN Weight Compression Framework Using Reweighted Optimization Methods
http://arxiv.org/abs/2004.05531
AUTHORS:        Tianyun Zhang ; Xiaolong Ma ; Zheng Zhan ; Shanglin Zhou ; Minghai Qin ; Fei Sun ; Yen-Kuang Chen ; Caiwen Ding ; Makan Fardad ; Yanzhi Wang
HIGHLIGHT:      In this paper, we propose a unified DNN weight pruning framework with dynamically updated regularization terms bounded by the designated constraint, which can generate both non-structured sparsity and different kinds of structured sparsity.

136, TITLE:     Online Initialization and Extrinsic Spatial-Temporal Calibration for Monocular Visual-Inertial Odometry
http://arxiv.org/abs/2004.05534
AUTHORS:        Weibo Huang ; Hong Liu ; Weiwei Wan
COMMENTS:       15 pages, 7 figures
HIGHLIGHT:      This paper presents an online initialization method for bootstrapping the optimization-based monocular visual-inertial odometry (VIO).

137, TITLE:     KD-MRI: A knowledge distillation framework for image reconstruction and image restoration in MRI workflow
http://arxiv.org/abs/2004.05319
AUTHORS:        Balamurali Murugesan ; Sricharan Vijayarangan ; Kaushik Sarveswaran ; Keerthi Ram ; Mohanasankar Sivaprakasam
COMMENTS:       Accepted in MIDL 2020. Code available
HIGHLIGHT:      In our work, we propose a knowledge distillation (KD) framework for the image to image problems in the MRI workflow in order to develop compact, low-parameter models without a significant drop in performance.

138, TITLE:     Improving Semantic Segmentation through Spatio-Temporal Consistency Learned from Videos
http://arxiv.org/abs/2004.05324
AUTHORS:        Ankita Pasad ; Ariel Gordon ; Tsung-Yi Lin ; Anelia Angelova
HIGHLIGHT:      We leverage unsupervised learning of depth, egomotion, and camera intrinsics to improve the performance of single-image semantic segmentation, by enforcing 3D-geometric and temporal consistency of segmentation masks across video frames.

139, TITLE:     Improving Disfluency Detection by Self-Training a Self-Attentive Model
http://arxiv.org/abs/2004.05323
AUTHORS:        Paria Jamshid Lou ; Mark Johnson
HIGHLIGHT:      Since the contextualized word embeddings are pre-trained on a large amount of unlabeled data, using additional unlabeled data to train a neural model might seem redundant.

140, TITLE:     FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions
http://arxiv.org/abs/2004.05565
AUTHORS:        Alvin Wan ; Xiaoliang Dai ; Peizhao Zhang ; Zijian He ; Yuandong Tian ; Saining Xie ; Bichen Wu ; Matthew Yu ; Tao Xu ; Kan Chen ; Peter Vajda ; Joseph E. Gonzalez
COMMENTS:       8 pages, 10 figures, accepted to CVPR 2020
HIGHLIGHT:      To address this bottleneck, we propose a memory and computationally efficient DNAS variant: DMaskingNAS.

141, TITLE:     Pre-training Text Representations as Meta Learning
http://arxiv.org/abs/2004.05568
AUTHORS:        Shangwen Lv ; Yuechen Wang ; Daya Guo ; Duyu Tang ; Nan Duan ; Fuqing Zhu ; Ming Gong ; Linjun Shou ; Ryan Ma ; Daxin Jiang ; Guihong Cao ; Ming Zhou ; Songlin Hu
COMMENTS:       2 figures, 3 tables
HIGHLIGHT:      In this work, we introduce a learning algorithm which directly optimizes model's ability to learn text representations for effective learning of downstream tasks.

142, TITLE:     Toward Hierarchical Self-Supervised Monocular Absolute Depth Estimation for Autonomous Driving Applications
http://arxiv.org/abs/2004.05560
AUTHORS:        Feng Xue ; Guirong Zhuo ; Ziyuan Huang ; Wufei Fu ; Zhuoyue Wu ; Marcelo H. Ang Jr
HIGHLIGHT:      In this work, we propose to address these two problems together by introducing DNet.

143, TITLE:     Multiparty Selection
http://arxiv.org/abs/2004.05548
AUTHORS:        Ke Chen ; Adrian Dumitrescu
COMMENTS:       11 pages, 2 figures
HIGHLIGHT:      Multiparty Selection

144, TITLE:     OpenMix: Reviving Known Knowledge for Discovering Novel Visual Categories in An Open World
http://arxiv.org/abs/2004.05551
AUTHORS:        Zhun Zhong ; Linchao Zhu ; Zhiming Luo ; Shaozi Li ; Yi Yang ; Nicu Sebe
HIGHLIGHT:      In this paper, we tackle the problem of discovering new classes in unlabeled visual data given labeled data from disjoint classes.

145, TITLE:     Rethinking Differentiable Search for Mixed-Precision Neural Networks
http://arxiv.org/abs/2004.05795
AUTHORS:        Zhaowei Cai ; Nuno Vasconcelos
COMMENTS:       accepted by CVPR 2020
HIGHLIGHT:      In this work, the problem of optimal mixed-precision network search (MPS) is considered.

146, TITLE:     Learning Event-Based Motion Deblurring
http://arxiv.org/abs/2004.05794
AUTHORS:        Zhe Jiang ; Yu Zhang ; Dongqing Zou ; Jimmy Ren ; Jiancheng Lv ; Yebin Liu
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      In this paper, we start from a sequential formulation of event-based motion deblurring, then show how its optimization can be unfolded with a novel end-to-end deep architecture.

147, TITLE:     Probabilistic Orientated Object Detection in Automotive Radar
http://arxiv.org/abs/2004.05310
AUTHORS:        Xu Dong ; Pengluo Wang ; Pengyue Zhang ; Langechuan Liu
COMMENTS:       Accepted at CVPR2020
HIGHLIGHT:      In this paper, we propose a deep-learning based algorithm for radar object detection. We created a new multimodal dataset with 102544 frames of raw radar and synchronized LiDAR data.

148, TITLE:     Feature Lenses: Plug-and-play Neural Modules for Transformation-Invariant Visual Representations
http://arxiv.org/abs/2004.05554
AUTHORS:        Shaohua Li ; Xiuchao Sui ; Jie Fu ; Yong Liu ; Rick Siow Mong Goh
COMMENTS:       20 pages
HIGHLIGHT:      To make CNNs more invariant to transformations, we propose "Feature Lenses", a set of ad-hoc modules that can be easily plugged into a trained model (referred to as the "host model").

149, TITLE:     Towards Transferable Adversarial Attack against Deep Face Recognition
http://arxiv.org/abs/2004.05790
AUTHORS:        Yaoyao Zhong ; Weihong Deng
HIGHLIGHT:      Extensive experiments on state-of-the-art face models with various training databases, loss functions and network architectures show that the proposed method can significantly enhance the transferability of existing attack methods. Finally, by applying DFANet to the LFW database, we generate a new set of adversarial face pairs that can successfully attack four commercial APIs without any queries.



==========Updates to Previous Papers==========
1, TITLE:       Reinforcement Learning in Healthcare: A Survey
http://arxiv.org/abs/1908.08796
AUTHORS:        Chao Yu ; Jiming Liu ; Shamim Nemati
HIGHLIGHT:      Such distinctive features make RL technique a suitable candidate for developing powerful solutions in a variety of healthcare domains, where diagnosing decisions or treatment regimes are usually characterized by a prolonged and sequential procedure.

2, TITLE:       One-Shot Texture Retrieval with Global Context Metric
http://arxiv.org/abs/1905.06656
AUTHORS:        Kai Zhu ; Wei Zhai ; Zheng-Jun Zha ; Yang Cao
COMMENTS:       ijcai2019-lastest
HIGHLIGHT:      In this paper, we tackle one-shot texture retrieval: given an example of a new reference texture, detect and segment all the pixels of the same texture category within an arbitrary image.

3, TITLE:       NAS-Bench-1Shot1: Benchmarking and Dissecting One-shot Neural Architecture Search
http://arxiv.org/abs/2001.10422
AUTHORS:        Arber Zela ; Julien Siems ; Frank Hutter
COMMENTS:       In: International Conference on Learning Representations (ICLR 2020); 19 pages, 17 figures
HIGHLIGHT:      In order to allow a scientific study of these components, we introduce a general framework for one-shot NAS that can be instantiated to many recently-introduced variants and introduce a general benchmarking framework that draws on the recent large-scale tabular benchmark NAS-Bench-101 for cheap anytime evaluations of one-shot NAS methods.

4, TITLE:       Person Re-identification in Aerial Imagery
http://arxiv.org/abs/1908.05024
AUTHORS:        Shizhou Zhang ; Qi Zhang ; Yifei Yang ; Xing Wei ; Peng Wang ; Bingliang Jiao ; Yanning Zhang
COMMENTS:       IEEE Transactions on Multimedia
HIGHLIGHT:      In this paper, to facilitate the research of person ReID in aerial imagery, we collect a large scale airborne person ReID dataset named as Person ReID for Aerial Imagery (PRAI-1581), which consists of 39,461 images of 1581 person identities.

5, TITLE:       HOnnotate: A method for 3D Annotation of Hand and Object Poses
http://arxiv.org/abs/1907.01481
AUTHORS:        Shreyas Hampali ; Mahdi Rad ; Markus Oberweger ; Vincent Lepetit
COMMENTS:       Accepted to CVPR2020
HIGHLIGHT:      We propose a method for annotating images of a hand manipulating an object with the 3D poses of both the hand and the object, together with a dataset created using this method.

6, TITLE:       Efficient Deep Representation Learning by Adaptive Latent Space Sampling
http://arxiv.org/abs/2004.02757
AUTHORS:        Yuanhan Mo ; Shuo Wang ; Chengliang Dai ; Rui Zhou ; Zhongzhao Teng ; Wenjia Bai ; Yike Guo
HIGHLIGHT:      To address the challenges of expensive annotations and loss of sample informativeness, here we propose a novel training framework which adaptively selects informative samples that are fed to the training process.

7, TITLE:       Robust Line Segments Matching via Graph Convolution Networks
http://arxiv.org/abs/2004.04993
AUTHORS:        QuanMeng Ma ; Guang Jiang ; DianZhi Lai
HIGHLIGHT:      In this paper, we present a new method of using a graph convolution network to match line segments in a pair of images, and we design a graph-based strategy of matching line segments with relaxing to an optimal transport problem.

8, TITLE:       Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstruction
http://arxiv.org/abs/2003.10983
AUTHORS:        Rohan Chabra ; Jan Eric Lenssen ; Eddy Ilg ; Tanner Schmidt ; Julian Straub ; Steven Lovegrove ; Richard Newcombe
HIGHLIGHT:      To address this problem we introduce Deep Local Shapes (DeepLS), a deep shape representation that enables encoding and reconstruction of high-quality 3D shapes without prohibitive memory requirements.

9, TITLE:       SpineNet: Learning Scale-Permuted Backbone for Recognition and Localization
http://arxiv.org/abs/1912.05027
AUTHORS:        Xianzhi Du ; Tsung-Yi Lin ; Pengchong Jin ; Golnaz Ghiasi ; Mingxing Tan ; Yin Cui ; Quoc V. Le ; Xiaodan Song
COMMENTS:       CVPR 2020
HIGHLIGHT:      In this paper, we argue encoder-decoder architecture is ineffective in generating strong multi-scale features because of the scale-decreased backbone.

10, TITLE:      Two-Stage Peer-Regularized Feature Recombination for Arbitrary Image Style Transfer
http://arxiv.org/abs/1906.02913
AUTHORS:        Jan Svoboda ; Asha Anoosheh ; Christian Osendorfer ; Jonathan Masci
HIGHLIGHT:      This paper introduces a neural style transfer model to generate a stylized image conditioning on a set of examples describing the desired style.

11, TITLE:      ColorFool: Semantic Adversarial Colorization
http://arxiv.org/abs/1911.10891
AUTHORS:        Ali Shahin Shamsabadi ; Ricardo Sanchez-Matilla ; Andrea Cavallaro
COMMENTS:       Conference on Computer Vision and Pattern Recognition (CVPR2020)
HIGHLIGHT:      In this paper, we propose a content-based black-box adversarial attack that generates unrestricted perturbations by exploiting image semantics to selectively modify colors within chosen ranges that are perceived as natural by humans.

12, TITLE:      Semi-supervised Image Attribute Editing using Generative Adversarial Networks
http://arxiv.org/abs/1907.01841
AUTHORS:        Yahya Dogan ; Hacer Yalim Keles
COMMENTS:       This paper is the preprint of the accepted manuscript in Neurocomputing Journal. To visualize the Figures in the manuscript in high quality, please check the version at this URL: https://github.com/yahyadogan72/CRG
HIGHLIGHT:      In this study, we introduce an architecture called Cyclic Reverse Generator (CRG), which allows learning the inverse function of the generator accurately via an encoder in an unsupervised setting by utilizing cyclic cost minimization.

13, TITLE:      MixNMatch: Multifactor Disentanglement and Encoding for Conditional Image Generation
http://arxiv.org/abs/1911.11758
AUTHORS:        Yuheng Li ; Krishna Kumar Singh ; Utkarsh Ojha ; Yong Jae Lee
COMMENTS:       CVPR 2020 camera ready
HIGHLIGHT:      We present MixNMatch, a conditional generative model that learns to disentangle and encode background, object pose, shape, and texture from real images with minimal supervision, for mix-and-match image generation.

14, TITLE:      Fast and Regularized Reconstruction of Building Façades from Street-View Images using Binary Integer Programming
http://arxiv.org/abs/2002.08549
AUTHORS:        Han Hu ; Libin Wang ; Mier Zhang ; Yulin Ding ; Qing Zhu
HIGHLIGHT:      Aiming to alleviate this issue, we cast the problem into binary integer programming, which omits the requirements for real value parameters and is more efficient to be solved .

15, TITLE:      COVID-MobileXpert: On-Device COVID-19 Screening using Snapshots of Chest X-Ray
http://arxiv.org/abs/2004.03042
AUTHORS:        Xin Li ; Chengyin Li ; Dongxiao Zhu
COMMENTS:       COVID-19, CoVid-19, SARS-CoV-2, on-device machine learning, Chest X-Ray (CXR)
HIGHLIGHT:      We present COVID-MobileXpert: a lightweight deep neural network (DNN) based mobile app that can use noisy snapshots of chest X-ray (CXR) for point-of-care COVID-19 screening.

16, TITLE:      Predicting Elastic Properties of Materials from Electronic Charge Density Using 3D Deep Convolutional Neural Networks
http://arxiv.org/abs/2003.13425
AUTHORS:        Yong Zhao ; Kunpeng Yuan ; Yinqiao Liu ; Steph-Yves Louis ; Ming Hu ; Jianjun Hu
COMMENTS:       15 pages; 5 figures
HIGHLIGHT:      Here, we propose to use electronic charge density (ECD) as a generic unified 3D descriptor for materials property prediction with the advantage of possessing close relation with the physical and chemical properties of materials.

17, TITLE:      Evolutionary Optimization of Deep Learning Activation Functions
http://arxiv.org/abs/2002.07224
AUTHORS:        Garrett Bingham ; William Macke ; Risto Miikkulainen
COMMENTS:       8 pages; 9 figures/tables; GECCO 2020
HIGHLIGHT:      This paper shows that evolutionary algorithms can discover novel activation functions that outperform ReLU.

18, TITLE:      Adversarial Attack and Defense on Point Sets
http://arxiv.org/abs/1902.10899
AUTHORS:        Jiancheng Yang ; Qiang Zhang ; Rongyao Fang ; Bingbing Ni ; Jinxian Liu ; Qi Tian
HIGHLIGHT:      Transferability of adversarial attacks between several point cloud networks is addressed, and we propose an momentum-enhanced pointwise gradient to improve the attack transferability.

19, TITLE:      Improved Basic Block Reordering
http://arxiv.org/abs/1809.04676
AUTHORS:        Andy Newell ; Sergey Pupyrev
COMMENTS:       Published in IEEE Transactions on Computers
HIGHLIGHT:      We propose a new algorithm that relies on a model combining the effects of fall-through and caching behavior.

20, TITLE:      TopoTag: A Robust and Scalable Topological Fiducial Marker System
http://arxiv.org/abs/1908.01450
AUTHORS:        Guoxing Yu ; Yongtao Hu ; Jingwen Dai
COMMENTS:       Accepted to TVCG
HIGHLIGHT:      Here we introduce TopoTag, a robust and scalable topological fiducial marker system, which supports reliable and accurate pose estimation from a single image. We collected a large test dataset including in total 169,713 images for evaluation, involving in-plane and out-of-plane rotation, image blur, different distances and various backgrounds, etc.

21, TITLE:      Theory III: Dynamics and Generalization in Deep Networks
http://arxiv.org/abs/1903.04991
AUTHORS:        Andrzej Banburski ; Qianli Liao ; Brando Miranda ; Lorenzo Rosasco ; Fernanda De La Torre ; Jack Hidary ; Tomaso Poggio
COMMENTS:       47 pages, 11 figures. This replaces previous versions of Theory III, that appeared on Arxiv [arXiv:1806.11379, arXiv:1801.00173] or on the CBMM site. v5: Changes throughout the paper to the presentation and tightening some of the statements
HIGHLIGHT:      We believe that the elusive complexity control we describe is responsible for the puzzling empirical finding of good predictive performance by deep networks, despite overparametrization.

22, TITLE:      Mining Domain Knowledge: Improved Framework towards Automatically Standardizing Anatomical Structure Nomenclature in Radiotherapy
http://arxiv.org/abs/1912.02084
AUTHORS:        Qiming Yang ; Hongyang Chao ; Dan Nguyen ; Steve Jiang
COMMENTS:       15 pages, 8 figures
HIGHLIGHT:      To solve these problems, we propose an automated structure nomenclature standardization framework, 3D Non-local Network with Voting (3DNNV).

23, TITLE:      A Comparison of the Taguchi Method and Evolutionary Optimization in Multivariate Testing
http://arxiv.org/abs/1808.08347
AUTHORS:        Jingbo Jiang ; Diego Legrand ; Robert Severn ; Risto Miikkulainen
COMMENTS:       5 pages, 4 figures, IAAI-19
HIGHLIGHT:      In contrast to the standard A/B testing, multivariate approach aims at evaluating a large number of values in a few key variables systematically.

24, TITLE:      IPGuard: Protecting Intellectual Property of Deep Neural Networks via Fingerprinting the Classification Boundary
http://arxiv.org/abs/1910.12903
AUTHORS:        Xiaoyu Cao ; Jinyuan Jia ; Neil Zhenqiang Gong
HIGHLIGHT:      In this work, we propose IPGuard, the first method to protect intellectual property of DNN classifiers that provably incurs no accuracy loss for the classifiers.

25, TITLE:      PANDA: Prototypical Unsupervised Domain Adaptation
http://arxiv.org/abs/2003.13274
AUTHORS:        Dapeng Hu ; Jian Liang ; Qibin Hou ; Hanshu Yan ; Yunpeng Chen ; Shuicheng Yan ; Jiashi Feng
HIGHLIGHT:      In this work, we attempt to calibrate the noisy pseudo labels with prototypes.

26, TITLE:      Attentive One-Dimensional Heatmap Regression for Facial Landmark Detection and Tracking
http://arxiv.org/abs/2004.02108
AUTHORS:        Shi Yin ; Shangfei Wang ; Xiaoping Chen ; Enhong Chen
HIGHLIGHT:      To address this, we propose a novel attentive one-dimensional heatmap regression method for facial landmark localization.

27, TITLE:      DR Loss: Improving Object Detection by Distributional Ranking
http://arxiv.org/abs/1907.10156
AUTHORS:        Qi Qian ; Lei Chen ; Hao Li ; Rong Jin
COMMENTS:       accepted by CVPR'20
HIGHLIGHT:      In this work, we propose a novel distributional ranking (DR) loss to handle the challenge.

28, TITLE:      Phase Portraits as Movement Primitives for Fast Humanoid Robot Control
http://arxiv.org/abs/1912.03535
AUTHORS:        Guilherme Maeda ; Okan Koc ; Jun Morimoto
HIGHLIGHT:      This article introduces Phase Portrait Movement Primitives (PPMP), a primitive that predicts dynamics on a low dimensional phase space which in turn is used to govern the high dimensional kinematics of the task.

29, TITLE:      Deep Representation Learning on Long-tailed Data: A Learnable Embedding Augmentation Perspective
http://arxiv.org/abs/2002.10826
AUTHORS:        Jialun Liu ; Yifan Sun ; Chuchu Han ; Zhaopeng Dou ; Wenhui Li
HIGHLIGHT:      To this end, we propose to construct each feature into a "feature cloud".

30, TITLE:      ARCH: Animatable Reconstruction of Clothed Humans
http://arxiv.org/abs/2004.04572
AUTHORS:        Zeng Huang ; Yuanlu Xu ; Christoph Lassner ; Hao Li ; Tony Tung
COMMENTS:       10 pages, 10 figures, CVPR2020
HIGHLIGHT:      In this paper, we propose ARCH (Animatable Reconstruction of Clothed Humans), a novel end-to-end framework for accurate reconstruction of animation-ready 3D clothed humans from a monocular image.

31, TITLE:      Category-wise Attack: Transferable Adversarial Examples for Anchor Free Object Detection
http://arxiv.org/abs/2003.04367
AUTHORS:        Quanyu Liao ; Xin Wang ; Bin Kong ; Siwei Lyu ; Youbing Yin ; Qi Song ; Xi Wu
HIGHLIGHT:      In this work, we aim to present an effective and efficient algorithm to generate adversarial examples to attack anchor-free object models based on two approaches.

32, TITLE:      EEG-Based Emotion Recognition Using Regularized Graph Neural Networks
http://arxiv.org/abs/1907.07835
AUTHORS:        Peixiang Zhong ; Di Wang ; Chunyan Miao
COMMENTS:       12 pages
HIGHLIGHT:      In this paper, we propose a regularized graph neural network (RGNN) for EEG-based emotion recognition.

33, TITLE:      Neural Reasoning, Fast and Slow, for Video Question Answering
http://arxiv.org/abs/1907.04553
AUTHORS:        Thao Minh Le ; Vuong Le ; Svetha Venkatesh ; Truyen Tran
HIGHLIGHT:      Inspired by the dual-process account of human reasoning, we design a dual process neural architecture, which is composed of a question-guided video processing module (System 1, fast and reactive) followed by a generic reasoning module (System 2, slow and deliberative).

34, TITLE:      IoU-aware Single-stage Object Detector for Accurate Localization
http://arxiv.org/abs/1912.05992
AUTHORS:        Shengkai Wu ; Xiaoping Li ; Xinggang Wang
HIGHLIGHT:      In this paper, IoU-aware single-stage object detector is proposed to solve this problem.

35, TITLE:      CTNN: Corticothalamic-inspired neural network
http://arxiv.org/abs/1910.12492
AUTHORS:        Leendert A Remmelzwaal ; Amit K Mishra ; George F R Ellis
HIGHLIGHT:      Here, we present a neural network architecture modelled on the top-down corticothalamic connections and the behaviour of the thalamus: a corticothalamic neural network (CTNN), consisting of an auto-encoder connected to a difference engine with a threshold.

36, TITLE:      Counting, uniform sampling, and enumeration of strings from non-deterministic finite automata
http://arxiv.org/abs/1906.09226
AUTHORS:        Marcelo Arenas ; Luis Alberto Croquevielle ; Rajesh Jayaram ; Cristian Riveros
COMMENTS:       To appear in PODS 2019
HIGHLIGHT:      In this work, we study two simple yet general complexity classes, based on logspace Turing machines, which provide a unifying framework for efficient query evaluation in areas like information extraction and graph databases, among others.

37, TITLE:      Enhanced POET: Open-Ended Reinforcement Learning through Unbounded Invention of Learning Challenges and their Solutions
http://arxiv.org/abs/2003.08536
AUTHORS:        Rui Wang ; Joel Lehman ; Aditya Rawal ; Jiale Zhi ; Yulun Li ; Jeff Clune ; Kenneth O. Stanley
COMMENTS:       23 pages, 14 figures
HIGHLIGHT:      Here we introduce and empirically validate two new innovations to the original algorithm, as well as two external innovations designed to help elucidate its full potential.

38, TITLE:      A Silver Standard Corpus of Human Phenotype-Gene Relations
http://arxiv.org/abs/1903.10728
AUTHORS:        Diana Sousa ; Andre Lamurias ; Francisco M. Couto
COMMENTS:       NAACL 2019
HIGHLIGHT:      This paper presents the Phenotype-Gene Relations (PGR) corpus, a silver standard corpus of human phenotype and gene annotations and their relations. We generated this corpus using Named-Entity Recognition tools, whose results were partially evaluated by eight curators, obtaining a precision of 87.01%.

39, TITLE:      Memory-Safety Challenge Considered Solved? An Empirical Study with All Rust CVEs
http://arxiv.org/abs/2003.03296
AUTHORS:        Hui Xu ; Zhuangbin Chen ; Mingshen Sun ; Yangfan Zhou
HIGHLIGHT:      The property is very attractive to developers, and many projects start using the language.

40, TITLE:      SAC: Accelerating and Structuring Self-Attention via Sparse Adaptive Connection
http://arxiv.org/abs/2003.09833
AUTHORS:        Xiaoya Li ; Yuxian Meng ; Mingxin Zhou ; Qinghong Han ; Fei Wu ; Jiwei Li
HIGHLIGHT:      In this paper, we present a method for accelerating and structuring self-attentions: Sparse Adaptive Connection (SAC).

41, TITLE:      Video Anomaly Detection for Smart Surveillance
http://arxiv.org/abs/2004.00222
AUTHORS:        Sijie Zhu ; Chen Chen ; Waqas Sultani
HIGHLIGHT:      In this paper, we provide a brief overview of the recent research progress on video anomaly detection and highlight a few future research directions.

42, TITLE:      SAM: The Sensitivity of Attribution Methods to Hyperparameters
http://arxiv.org/abs/2003.08754
AUTHORS:        Naman Bansal ; Chirag Agarwal ; Anh Nguyen
COMMENTS:       Oral paper at CVPR 2020
HIGHLIGHT:      In this paper, we provide a thorough empirical study on the sensitivity of existing attribution methods.

43, TITLE:      Optimal Farsighted Agents Tend to Seek Power
http://arxiv.org/abs/1912.01683
AUTHORS:        Alexander Matt Turner ; Logan Smith ; Rohin Shah ; Prasad Tadepalli
COMMENTS:       8 pages, 25 pages with appendices
HIGHLIGHT:      Others have voiced skepticism: humans seem idiosyncratic in their urges to power, which need not be present in the agents we design.

44, TITLE:      COVID_MTNet: COVID-19 Detection with Multi-Task Deep Learning Approaches
http://arxiv.org/abs/2004.03747
AUTHORS:        Md Zahangir Alom ; M M Shaifur Rahman ; Mst Shamima Nasrin ; Tarek M. Taha ; Vijayan K. Asari
COMMENTS:       11 pages, 15 figures
HIGHLIGHT:      In this paper, we propose a fast and efficient way to identify COVID-19 patients with multi-task deep learning (DL) methods.

45, TITLE:      MRQy: An Open-Source Tool for Quality Control of MR Imaging Data
http://arxiv.org/abs/2004.04871
AUTHORS:        Amir Reza Sadri ; Andrew Janowczyk ; Ren Zou ; Ruchika Verma ; Jacob Antunes ; Anant Madabhushi ; Pallavi Tiwari ; Satish E. Viswanath
COMMENTS:       23 pages, 7 figures. Submitted to Medical Physics
HIGHLIGHT:      We present MRQy, a new open-source quality control tool to (a) interrogate MRI cohorts for site- or equipment-based differences, and (b) quantify the impact of MRI artifacts on relative image quality; to help determine how to correct for these variations prior to model development.

46, TITLE:      Where Does It End? -- Reasoning About Hidden Surfaces by Object Intersection Constraints
http://arxiv.org/abs/2004.04630
AUTHORS:        Michael Strecke ; Joerg Stueckler
COMMENTS:       Accepted to the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
HIGHLIGHT:      In this paper we propose Co-Section, an optimization-based approach to 3D dynamic scene reconstruction, which infers hidden shape information from intersection constraints.

47, TITLE:      Pyramid Focusing Network for mutation prediction and classification in CT images
http://arxiv.org/abs/2004.03302
AUTHORS:        Xukun Zhang ; Wenxin Hu ; Wen Wu
COMMENTS:       The carelessness of writing the paper led to the incorrect experimental results shown in the paper.In addition, the experimental design shown in the paper is incomplete, and I need to revise it substantially before submitting it
HIGHLIGHT:      In this paper, we propose a pyramid focusing network (PFNet) for mutation prediction and classification based on CT images.

48, TITLE:      White-Box Evaluation of Fingerprint Matchers: Robustness to Minutiae Perturbations
http://arxiv.org/abs/1909.00799
AUTHORS:        Steven A. Grosz ; Joshua J. Engelsma ; Nicholas G. Paulter Jr. ; Anil K. Jain
HIGHLIGHT:      We report results of a controlled, white-box evaluation of one open-source and two commercial-off-the-shelf (COTS) minutiae-based matchers in terms of their robustness against controlled perturbations (random noise and non-linear distortions) introduced into the input minutiae feature sets.

49, TITLE:      A Survey on Contextual Embeddings
http://arxiv.org/abs/2003.07278
AUTHORS:        Qi Liu ; Matt J. Kusner ; Phil Blunsom
COMMENTS:       13 pages
HIGHLIGHT:      In this survey, we review existing contextual embedding models, cross-lingual polyglot pre-training, the application of contextual embeddings in downstream tasks, model compression, and model analyses.

50, TITLE:      A New Dataset for Natural Language Inference from Code-mixed Conversations
http://arxiv.org/abs/2004.05051
AUTHORS:        Simran Khanuja ; Sandipan Dandapat ; Sunayana Sitaram ; Monojit Choudhury
COMMENTS:       To appear in CALCS, LREC 2020
HIGHLIGHT:      In this paper, we present the first dataset for code-mixed NLI, in which both the premises and hypotheses are in code-mixed Hindi-English.

51, TITLE:      Generator evaluator-selector net for panoptic image segmentation and splitting unfamiliar objects into parts
http://arxiv.org/abs/1908.09108
AUTHORS:        Sagi Eppel ; Alan Aspuru-Guzik
HIGHLIGHT:      In machine learning and other fields, suggesting a good solution to a problem is usually a harder task than evaluating the quality of such a solution.

52, 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.

53, TITLE:      A Neuromorphic Proto-Object Based Dynamic Visual Saliency Model with an FPGA Implementation
http://arxiv.org/abs/2002.11898
AUTHORS:        Jamal Lottier Molin ; Chetan Singh Thakur ; Ralph Etienne-Cummings ; Ernst Niebur
COMMENTS:       15 pages, 6 figures, 6 tables, journal
HIGHLIGHT:      To address this, we introduce a Field-Programmable Gate Array implementation of the model on an Opal Kelly 7350 Kintex-7 board.

54, TITLE:      EvolveGraph: Heterogeneous Multi-Agent Multi-Modal Trajectory Prediction with Evolving Interaction Graphs
http://arxiv.org/abs/2003.13924
AUTHORS:        Jiachen Li ; Fan Yang ; Masayoshi Tomizuka ; Chiho Choi
COMMENTS:       19 pages
HIGHLIGHT:      In this paper, we propose a generic trajectory forecasting framework (named EvolveGraph) with explicit interaction modeling via a latent interaction graph among multiple heterogeneous, interactive agents.

55, TITLE:      Early Forecasting of Text Classification Accuracy and F-Measure with Active Learning
http://arxiv.org/abs/2001.10337
AUTHORS:        Thomas Orth ; Michael Bloodgood
COMMENTS:       8 pages, 9 figures, 2 tables; published in Proceedings of the IEEE 14th International Conference on Semantic Computing (ICSC), San Diego, CA, USA, pages 77-84, February 2020
HIGHLIGHT:      We investigate the difference in forecasting difficulty when using accuracy and F-measure as the text classification system performance metrics and we find that F-measure is more difficult to forecast.

56, TITLE:      Fast-Learning and High-Robustness Unsupervised Spiking Neural Networks with a Highly Parallel Inception-like Architecture
http://arxiv.org/abs/2001.01680
AUTHORS:        Mingyuan Meng ; Xingyu Yang ; Shanlin Xiao ; Zhiyi Yu
HIGHLIGHT:      In this paper, to ease these limitations: 1) We designed a highly parallel multi-pathway network architecture, inspired by the Inception module in the Artificial Neural Network (ANN) literature.

57, TITLE:      End-to-end training of time domain audio separation and recognition
http://arxiv.org/abs/1912.08462
AUTHORS:        Thilo von Neumann ; Keisuke Kinoshita ; Lukas Drude ; Christoph Boeddeker ; Marc Delcroix ; Tomohiro Nakatani ; Reinhold Haeb-Umbach
COMMENTS:       5 pages, 1 figure, to appear in ICASSP 2020
HIGHLIGHT:      We here demonstrate how to combine a separation module based on a Convolutional Time domain Audio Separation Network (Conv-TasNet) with an E2E speech recognizer and how to train such a model jointly by distributing it over multiple GPUs or by approximating truncated back-propagation for the convolutional front-end.

58, TITLE:      Domain Agnostic Feature Learning for Image and Video Based Face Anti-spoofing
http://arxiv.org/abs/1912.07124
AUTHORS:        Suman Saha ; Wenhao Xu ; Menelaos Kanakis ; Stamatios Georgoulis ; Yuhua Chen ; Danda Pani Paudel ; Luc Van Gool
COMMENTS:       CVPR 2020 Biometrics Workshop
HIGHLIGHT:      In this paper, we explicitly tackle this problem and propose a class-conditional domain discriminator module, that, coupled with a gradient reversal layer, tries to generate live and spoof features that are discriminative, but at the same time robust against the aforementioned variability factors.

59, TITLE:      Meta-Learning Initializations for Image Segmentation
http://arxiv.org/abs/1912.06290
AUTHORS:        Sean M. Hendryx ; Andrew B. Leach ; Paul D. Hein ; Clayton T. Morrison
HIGHLIGHT:      In this work, we study first-order meta-learning of initializations for deep neural networks that must produce dense, structured predictions given an arbitrary amount of training data for a new task.

60, TITLE:      Nonuniform Families of Polynomial-Size Quantum Finite Automata and Quantum Logarithmic-Space Computation with Polynomial-Size Advice
http://arxiv.org/abs/1907.02916
AUTHORS:        Tomoyuki Yamakami
COMMENTS:       (10pt, A4, pp.29, 1 figure) This current version corrects and extends the preliminary version that appeared in the Proceedings of the 13th International Conference on Language and Automata Theory and Applications (LATA 2019), Saint Petersburg, Russia, March 26-29, 2019, Lecture Notes in Computer Science, Springer, vol. 11417, pp. 134-145, 2019
HIGHLIGHT:      In a similar fashion, we introduce nonuniform state complexity classes using nonuniform families of quantum finite automata empowered by the flexible use of garbage tapes.

61, TITLE:      AraNet: A Deep Learning Toolkit for Arabic Social Media
http://arxiv.org/abs/1912.13072
AUTHORS:        Muhammad Abdul-Mageed ; Chiyu Zhang ; Azadeh Hashemi ; El Moatez Billah Nagoudi
COMMENTS:       Accepted by The 4th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT)
HIGHLIGHT:      We describe AraNet, a collection of deep learning Arabic social media processing tools.

62, TITLE:      Learning Representations For Images With Hierarchical Labels
http://arxiv.org/abs/2004.00909
AUTHORS:        Ankit Dhall
COMMENTS:       Master thesis
HIGHLIGHT:      In this thesis we present a set of methods to leverage information about the semantic hierarchy induced by class labels.

63, TITLE:      Entity Abstraction in Visual Model-Based Reinforcement Learning
http://arxiv.org/abs/1910.12827
AUTHORS:        Rishi Veerapaneni ; John D. Co-Reyes ; Michael Chang ; Michael Janner ; Chelsea Finn ; Jiajun Wu ; Joshua B. Tenenbaum ; Sergey Levine
COMMENTS:       Accepted at CoRL 2019
HIGHLIGHT:      We present object-centric perception, prediction, and planning (OP3), which to the best of our knowledge is the first fully probabilistic entity-centric dynamic latent variable framework for model-based reinforcement learning that acquires entity representations from raw visual observations without supervision and uses them to predict and plan.

64, TITLE:      The role of formalism in system requirements (extended version)
http://arxiv.org/abs/1911.02564
AUTHORS:        Jean-Michel Bruel ; Sophie Ebersold ; Florian Galinier ; Alexandr Naumchev ; Manuel Mazzara ; Bertrand Meyer
COMMENTS:       Third version (7 April 2020)
HIGHLIGHT:      The present survey discusses some of the main formal approaches and compares them to informal methods.

65, TITLE:      Sketch Less for More: On-the-Fly Fine-Grained Sketch Based Image Retrieval
http://arxiv.org/abs/2002.10310
AUTHORS:        Ayan Kumar Bhunia ; Yongxin Yang ; Timothy M. Hospedales ; Tao Xiang ; Yi-Zhe Song
COMMENTS:       IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2020 [Oral Presentation] Code: https://github.com/AyanKumarBhunia/on-the-fly-FGSBIR
HIGHLIGHT:      In this paper, we reformulate the conventional FG-SBIR framework to tackle these challenges, with the ultimate goal of retrieving the target photo with the least number of strokes possible.

66, TITLE:      BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networks
http://arxiv.org/abs/1911.10127
AUTHORS:        Yao Yao ; Zixin Luo ; Shiwei Li ; Jingyang Zhang ; Yufan Ren ; Lei Zhou ; Tian Fang ; Long Quan
COMMENTS:       Accepted to CVPR2020
HIGHLIGHT:      In this paper, we introduce BlendedMVS, a novel large-scale dataset, to provide sufficient training ground truth for learning-based MVS.

67, TITLE:      Multi-Task Learning of Generalizable Representations for Video Action Recognition
http://arxiv.org/abs/1811.08362
AUTHORS:        Zhiyu Yao ; Yunbo Wang ; Mingsheng Long ; Jianmin Wang ; Philip S Yu ; Jiaguang Sun
COMMENTS:       ICME 2020
HIGHLIGHT:      Based on these findings, we present a multi-task learning paradigm for video classification.

68, TITLE:      Revisiting Training Strategies and Generalization Performance in Deep Metric Learning
http://arxiv.org/abs/2002.08473
AUTHORS:        Karsten Roth ; Timo Milbich ; Samarth Sinha ; Prateek Gupta ; Björn Ommer ; Joseph Paul Cohen
HIGHLIGHT:      Exploiting these insights, we propose a simple, yet effective, training regularization to reliably boost the performance of ranking-based DML models on various standard benchmark datasets.

69, TITLE:      To See in the Dark: N2DGAN for Background Modeling in Nighttime Scene
http://arxiv.org/abs/1912.06556
AUTHORS:        Zhenfeng Zhu ; Yingying Meng ; Deqiang Kong ; Xingxing Zhang ; Yandong Guo ; Yao Zhao
COMMENTS:       11 pages, 11 figures. The paper has been accepted by IEEE Transactions on Circuits and Systems for Video Technology(TCSVT)
HIGHLIGHT:      Specifically, to balance the preservation of background scene and the foreground object(s) in generating the virtual daytime image, we present a two-pathway generation model, in which the global and local sub-networks are well combined with spatial and temporal consistency constraints.

70, TITLE:      Knowledge Guided Text Retrieval and Reading for Open Domain Question Answering
http://arxiv.org/abs/1911.03868
AUTHORS:        Sewon Min ; Danqi Chen ; Luke Zettlemoyer ; Hannaneh Hajishirzi
HIGHLIGHT:      We introduce an approach for open-domain question answering (QA) that retrieves and reads a passage graph, where vertices are passages of text and edges represent relationships that are derived from an external knowledge base or co-occurrence in the same article.

71, TITLE:      Mimic and Fool: A Task Agnostic Adversarial Attack
http://arxiv.org/abs/1906.04606
AUTHORS:        Akshay Chaturvedi ; Utpal Garain
HIGHLIGHT:      Keeping this in mind, we propose Mimic and Fool, a task agnostic adversarial attack.

72, TITLE:      Long-tail learning with attributes
http://arxiv.org/abs/2004.02235
AUTHORS:        Dvir Samuel ; Yuval Atzmon ; Gal Chechik
HIGHLIGHT:      Here we describe DRAGON, a novel modular architecture for long-tail learning designed to address these biases and fuse multi-modal information in face of unbalanced data. We present the first benchmark for long-tail learning with attributes and use it to evaluate DRAGON.

73, TITLE:      How Often Do Single-Statement Bugs Occur? The ManySStuBs4J Dataset
http://arxiv.org/abs/1905.13334
AUTHORS:        Rafael-Michael Karampatsis ; Charles Sutton
COMMENTS:       5 pages; to appear in Proceedings of MSR 2020
HIGHLIGHT:      In an initial analysis, we find that about 33% of the simple bug fixes match the templates, indicating that a remarkable number of single-statement bugs can be repaired with a relatively small set of templates. To fill this gap, we provide a dataset of 153,652 single statement bug-fix changes mined from 1,000 popular open-source Java projects, annotated by whether they match any of a set of 16 bug templates, inspired by state-of-the-art program repair techniques.

74, TITLE:      Talk to Papers: Bringing Neural Question Answering to Academic Search
http://arxiv.org/abs/2004.02002
AUTHORS:        Tianchang Zhao ; Kyusong Lee
COMMENTS:       demo paper accepted at ACL 2020
HIGHLIGHT:      We introduce Talk to Papers, which exploits the recent open-domain question answering (QA) techniques to improve the current experience of academic search.

75, TITLE:      On-Device Text Representations Robust To Misspellings via Projections
http://arxiv.org/abs/1908.05763
AUTHORS:        Chinnadhurai Sankar ; Sujith Ravi ; Zornitsa Kozareva
HIGHLIGHT:      In this paper, we show that the projection based neural classifiers are inherently robust to misspellings and perturbations of the input text.

76, TITLE:      CvxNet: Learnable Convex Decomposition
http://arxiv.org/abs/1909.05736
AUTHORS:        Boyang Deng ; Kyle Genova ; Soroosh Yazdani ; Sofien Bouaziz ; Geoffrey Hinton ; Andrea Tagliasacchi
HIGHLIGHT:      We introduce a network architecture to represent a low dimensional family of convexes.

77, TITLE:      Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks
http://arxiv.org/abs/1910.01279
AUTHORS:        Haofan Wang ; Zifan Wang ; Mengnan Du ; Fan Yang ; Zijian Zhang ; Sirui Ding ; Piotr Mardziel ; Xia Hu
COMMENTS:       Accepted to CVPR 2020: Workshop on Fair, Data Efficient and Trusted Computer Vision
HIGHLIGHT:      In this paper, we develop a novel post-hoc visual explanation method called Score-CAM based on class activation mapping.

78, TITLE:      Inclusive GAN: Improving Data and Minority Coverage in Generative Models
http://arxiv.org/abs/2004.03355
AUTHORS:        Ning Yu ; Ke Li ; Peng Zhou ; Jitendra Malik ; Larry Davis ; Mario Fritz
HIGHLIGHT:      We develop an extension that allows explicit control over the minority subgroups that the model should ensure to include, and validate its effectiveness at little compromise from the overall performance on the entire dataset.

79, TITLE:      CenterMask: single shot instance segmentation with point representation
http://arxiv.org/abs/2004.04446
AUTHORS:        Yuqing Wang ; Zhaoliang Xu ; Hao Shen ; Baoshan Cheng ; Lirong Yang
COMMENTS:       To appear at CVPR 2020
HIGHLIGHT:      In this paper, we propose a single-shot instance segmentation method, which is simple, fast and accurate.

80, TITLE:      Encoder-Decoder Based Convolutional Neural Networks with Multi-Scale-Aware Modules for Crowd Counting
http://arxiv.org/abs/2003.05586
AUTHORS:        Pongpisit Thanasutives ; Ken-ichi Fukui ; Masayuki Numao ; Boonserm Kijsirikul
COMMENTS:       Update the results and correct typo
HIGHLIGHT:      In this paper, we proposed two modified neural network architectures based on SFANet and SegNet respectively for accurate and efficient crowd counting.

81, TITLE:      Procedure Planning in Instructional Videos
http://arxiv.org/abs/1907.01172
AUTHORS:        Chien-Yi Chang ; De-An Huang ; Danfei Xu ; Ehsan Adeli ; Li Fei-Fei ; Juan Carlos Niebles
COMMENTS:       14 pages, 7 figures
HIGHLIGHT:      In this paper, we study the problem of procedure planning in instructional videos, which can be seen as a step towards enabling autonomous agents to plan for complex tasks in everyday settings such as cooking.

82, TITLE:      Synthesizing Structured CAD Models with Equality Saturation and Inverse Transformations
http://arxiv.org/abs/1909.12252
AUTHORS:        Chandrakana Nandi ; Max Willsey ; Adam Anderson ; James R. Wilcox ; Eva Darulova ; Dan Grossman ; Zachary Tatlock
COMMENTS:       14 pages
HIGHLIGHT:      We present Szalinski, a tool that uses Equality Saturation with semantics-preserving CAD rewrites to efficiently search for smaller equivalent programs.