2020.04.15.txt

==========New Papers==========
1, TITLE:       The Devil is in the Details: Self-Supervised Attention for Vehicle Re-Identification
http://arxiv.org/abs/2004.06271
AUTHORS:        Pirazh Khorramshahi ; Neehar Peri ; Jun-cheng Chen ; Rama Chellappa
HIGHLIGHT:      In this paper, we present Self-supervised Attention for Vehicle Re-identification (SAVER), a novel approach to effectively learn vehicle-specific discriminative features.

2, TITLE:       Bidirectional Graph Reasoning Network for Panoptic Segmentation
http://arxiv.org/abs/2004.06272
AUTHORS:        Yangxin Wu ; Gengwei Zhang ; Yiming Gao ; Xiajun Deng ; Ke Gong ; Xiaodan Liang ; Liang Lin
COMMENTS:       CVPR2020
HIGHLIGHT:      We introduce a Bidirectional Graph Reasoning Network (BGRNet), which incorporates graph structure into the conventional panoptic segmentation network to mine the intra-modular and intermodular relations within and between foreground things and background stuff classes.

3, TITLE:       RealMonoDepth: Self-Supervised Monocular Depth Estimation for General Scenes
http://arxiv.org/abs/2004.06267
AUTHORS:        Mertalp Ocal ; Armin Mustafa
HIGHLIGHT:      We present a generalised self-supervised learning approach for monocular estimation of the real depth across scenes with diverse depth ranges from 1--100s of meters.

4, TITLE:       Smart Inference for Multidigit Convolutional Neural Network based Barcode Decoding
http://arxiv.org/abs/2004.06297
AUTHORS:        Thao Do ; Yalew Tolcha ; Tae Joon Jun ; Daeyoung Kim
HIGHLIGHT:      This work aims to solve the decoding problem using deep convolutional neural network with the possibility of running on portable devices. Secondly, we have created a large practical evaluation dataset of real captured 1D barcode under various challenging conditions to test our methods vigorously, which is publicly available for other researchers.

5, TITLE:       Cross-Lingual Semantic Role Labeling with High-Quality Translated Training Corpus
http://arxiv.org/abs/2004.06295
AUTHORS:        Hao Fei ; Meishan Zhang ; Donghong Ji
COMMENTS:       ACL 2020
HIGHLIGHT:      In this paper, we propose a novel alternative based on corpus translation, constructing high-quality training datasets for the target languages from the source gold-standard SRL annotations.

6, TITLE:       Deep Entwined Learning Head Pose and Face Alignment Inside an Attentional Cascade with Doubly-Conditional fusion
http://arxiv.org/abs/2004.06558
AUTHORS:        Arnaud Dapogny ; Kévin Bailly ; Matthieu Cord
COMMENTS:       Accepted for publication as an oral session @IEEE FG2020
HIGHLIGHT:      In this paper, we propose to entwine face alignment and head pose tasks inside an attentional cascade.

7, TITLE:       Multi-Ontology Refined Embeddings (MORE): A Hybrid Multi-Ontology and Corpus-based Semantic Representation for Biomedical Concepts
http://arxiv.org/abs/2004.06555
AUTHORS:        Steven Jiang ; Weiyi Wu ; Naofumi Tomita ; Craig Ganoe ; Saeed Hassanpour
HIGHLIGHT:      This paper introduces Multi-Ontology Refined Embeddings (MORE), a novel hybrid framework for incorporating domain knowledge from multiple ontologies into a distributional semantic model, learned from a corpus of clinical text.

8, TITLE:       Unsupervised Performance Analysis of 3D Face Alignment
http://arxiv.org/abs/2004.06550
AUTHORS:        Mostafa Sadeghi ; Sylvain Guy ; Adrien Raison ; Xavier Alameda-Pineda ; Radu Horaud
HIGHLIGHT:      We revisit the problem of robust estimation of the rigid transformation between two point sets and we describe two algorithms, one based on a mixture between a Gaussian and a uniform distribution, and another one based on the generalized Student's t-distribution.

9, TITLE:       Fast Mutation in Crossover-based Algorithms
http://arxiv.org/abs/2004.06538
AUTHORS:        Denis Antipov ; Maxim Buzdalov ; Benjamin Doerr
COMMENTS:       This is a version of the same paper presented at GECCO 2020 completed with the proofs which were missing because of the page limit
HIGHLIGHT:      In this first runtime analysis of a crossover-based algorithm using a heavy-tailed choice of the mutation rate, we show an even stronger impact.

10, TITLE:      VehicleNet: Learning Robust Visual Representation for Vehicle Re-identification
http://arxiv.org/abs/2004.06305
AUTHORS:        Zhedong Zheng ; Tao Ruan ; Yunchao Wei ; Yi Yang ; Tao Mei
HIGHLIGHT:      As the existing vehicle datasets are limited in terms of training images and viewpoints, we propose to build a unique large-scale vehicle dataset (called VehicleNet) by harnessing four public vehicle datasets, and design a simple yet effective two-stage progressive approach to learning more robust visual representation from VehicleNet.

11, TITLE:      Few-Shot Single-View 3-D Object Reconstruction with Compositional Priors
http://arxiv.org/abs/2004.06302
AUTHORS:        Mateusz Michalkiewicz ; Sarah Parisot ; Stavros Tsogkas ; Mahsa Baktashmotlagh ; Anders Eriksson ; Eugene Belilovsky
HIGHLIGHT:      In this work we demonstrate experimentally that naive baselines do not apply when the goal is to learn to reconstruct novel objects using very few examples, and that in a \emph{few-shot} learning setting, the network must learn concepts that can be applied to new categories, avoiding rote memorization.

12, TITLE:      Quantifying Community Characteristics of Maternal Mortality Using Social Media
http://arxiv.org/abs/2004.06303
AUTHORS:        Rediet Abebe ; Salvatore Giorgi ; Anna Tedijanto ; Anneke Buffone ; H. Andrew Schwartz
COMMENTS:       In Proceedings of The Web Conference 2020(WWW '20)
HIGHLIGHT:      In this work, we explore the role that social media language can play in providing insights into such community characteristics.

13, TITLE:      Multilingual Machine Translation: Closing the Gap between Shared and Language-specific Encoder-Decoders
http://arxiv.org/abs/2004.06575
AUTHORS:        Carlos Escolano ; Marta R. Costa-jussà ; José A. R. Fonollosa ; Mikel Artetxe
HIGHLIGHT:      In this paper, we propose an alternative approach that is based on language-specific encoder-decoders, and can thus be more easily extended to new languages by learning their corresponding modules.

14, TITLE:      WQT and DG-YOLO: towards domain generalization in underwater object detection
http://arxiv.org/abs/2004.06333
AUTHORS:        Hong Liu ; Pinhao Song ; Runwei Ding
HIGHLIGHT:      This paper aims to build a GUOD with small underwater dataset with limited types of water quality.

15, TITLE:      Automated Diabetic Retinopathy Grading using Deep Convolutional Neural Network
http://arxiv.org/abs/2004.06334
AUTHORS:        Saket S. Chaturvedi ; Kajol Gupta ; Vaishali Ninawe ; Prakash S. Prasad
COMMENTS:       \c{opyright} 20xx 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:      In this study, we have utilized a pre-trained DenseNet121 network with several modifications and trained on APTOS 2019 dataset.

16, TITLE:      Transformer based Grapheme-to-Phoneme Conversion
http://arxiv.org/abs/2004.06338
AUTHORS:        Sevinj Yolchuyeva ; Géza Németh ; Bálint Gyires-Tóth
COMMENTS:       INTERSPEECH 2019
HIGHLIGHT:      In this paper, we investigate the application of transformer architecture to G2P conversion and compare its performance with recurrent and convolutional neural network based approaches.

17, TITLE:      Transfer Learning with Deep Convolutional Neural Network (CNN) for Pneumonia Detection using Chest X-ray
http://arxiv.org/abs/2004.06578
AUTHORS:        Tawsifur Rahman ; Muhammad E. H. Chowdhury ; Amith Khandakar ; Khandaker R. Islam ; Khandaker F. Islam ; Zaid B. Mahbub ; Muhammad A. Kadir ; Saad Kashem
COMMENTS:       13 Figures, 5 tables. arXiv admin note: text overlap with arXiv:2003.13145
HIGHLIGHT:      The aim of this paper is to automatically detect bacterial and viral pneumonia using digital x-ray images.

18, TITLE:      dMFEA-II: An Adaptive Multifactorial Evolutionary Algorithm for Permutation-based Discrete Optimization Problems
http://arxiv.org/abs/2004.06559
AUTHORS:        Eneko Osaba ; Aritz D. Martinez ; Akemi Galvez ; Andres Iglesias ; Javier Del Ser
COMMENTS:       7 pages, 0 figures, submitted to ECPERM - Evolutionary Computation for Permutation Problems workshop, part of The Genetic and Evolutionary Computation Conference 2020 (GECCO 2020),
HIGHLIGHT:      In this paper we entirely reformulate such concepts, making them suited to deal with permutation-based search spaces without loosing the inherent benefits of MFEA-II.

19, TITLE:      A Tailored NSGA-III Instantiation for Flexible Job Shop Scheduling
http://arxiv.org/abs/2004.06564
AUTHORS:        Yali Wang ; Bas van Stein ; Michael T. M. Emmerich ; Thomas Bäck
HIGHLIGHT:      A customized multi-objective evolutionary algorithm (MOEA) is proposed for the multi-objective flexible job shop scheduling problem (FJSP).

20, TITLE:      A2D2: Audi Autonomous Driving Dataset
http://arxiv.org/abs/2004.06320
AUTHORS:        Jakob Geyer ; Yohannes Kassahun ; Mentar Mahmudi ; Xavier Ricou ; Rupesh Durgesh ; Andrew S. Chung ; Lorenz Hauswald ; Viet Hoang Pham ; Maximilian Mühlegg ; Sebastian Dorn ; Tiffany Fernandez ; Martin Jänicke ; Sudesh Mirashi ; Chiragkumar Savani ; Martin Sturm ; Oleksandr Vorobiov ; Martin Oelker ; Sebastian Garreis ; Peter Schuberth
COMMENTS:       https://www.a2d2.audi/
HIGHLIGHT:      To this end, we release the Audi Autonomous Driving Dataset (A2D2).

21, TITLE:      Counting Small Induced Subgraphs Satisfying Monotone Properties
http://arxiv.org/abs/2004.06595
AUTHORS:        Marc Roth ; Johannes Schmitt ; Philip Wellnitz
COMMENTS:       33 pages, 2 figures
HIGHLIGHT:      In this work, we fully answer and explicitly classify the case of monotone, that is subgraph-closed, properties: We show that for any non-trivial monotone property $\Phi$, the problem $\#\mathsf{IndSub}(\Phi)$ cannot be solved in time $f(k)\cdot |V(G)|^{o(k/ {\log^{1/2}(k)})}$ for any function $f$, unless the Exponential Time Hypothesis fails.

22, TITLE:      Knowledge Elicitation using Deep Metric Learning and Psychometric Testing
http://arxiv.org/abs/2004.06353
AUTHORS:        Lu Yin ; Vlado Menkovski ; Mykola Pechenizkiy
COMMENTS:       16 pages, 11 figures
HIGHLIGHT:      In this paper, we provide a method for efficient hierarchical knowledge elicitation (HKE) from experts working with high-dimensional data such as images or videos.

23, TITLE:      Speech Translation and the End-to-End Promise: Taking Stock of Where We Are
http://arxiv.org/abs/2004.06358
AUTHORS:        Matthias Sperber ; Matthias Paulik
COMMENTS:       ACL 2020 theme track
HIGHLIGHT:      This paper provides a unifying categorization and nomenclature that covers both traditional and recent approaches and that may help researchers by highlighting both trade-offs and open research questions.

24, TITLE:      InsideBias: Measuring Bias in Deep Networks and Application to Face Gender Biometrics
http://arxiv.org/abs/2004.06592
AUTHORS:        Ignacio Serna ; Alejandro Peña ; Aythami Morales ; Julian Fierrez
HIGHLIGHT:      We present a comprehensive analysis of bias effects when using an unbalanced training dataset on the features learned by the models.

25, TITLE:      Walk the Lines: Object Contour Tracing CNN for Contour Completion of Ships
http://arxiv.org/abs/2004.06587
AUTHORS:        André Peter Kelm ; Udo Zölzer
COMMENTS:       Submission to the ICPR
HIGHLIGHT:      We develop a new contour tracing algorithm to enhance the results of the latest object contour detectors.

26, TITLE:      Code Completion using Neural Attention and Byte Pair Encoding
http://arxiv.org/abs/2004.06343
AUTHORS:        Youri Arkesteijn ; Nikhil Saldanha ; Bastijn Kostense
COMMENTS:       4 pages, 4 figures, 1 table
HIGHLIGHT:      In this paper, we aim to do code completion based on implementing a Neural Network from Li et.

27, TITLE:      Stochastic batch size for adaptive regularization in deep network optimization
http://arxiv.org/abs/2004.06341
AUTHORS:        Kensuke Nakamura ; Stefano Soatto ; Byung-Woo Hong
HIGHLIGHT:      We propose a first-order stochastic optimization algorithm incorporating adaptive regularization applicable to machine learning problems in deep learning framework.

28, TITLE:      A Primal-Dual Solver for Large-Scale Tracking-by-Assignment
http://arxiv.org/abs/2004.06375
AUTHORS:        Stefan Haller ; Mangal Prakash ; Lisa Hutschenreiter ; Tobias Pietzsch ; Carsten Rother ; Florian Jug ; Paul Swoboda ; Bogdan Savchynskyy
COMMENTS:       23rd International Conference on Artificial Intelligence and Statistics (AISTATS), 2020
HIGHLIGHT:      We propose a fast approximate solver for the combinatorial problem known as tracking-by-assignment, which we apply to cell tracking.

29, TITLE:      Footprints and Free Space from a Single Color Image
http://arxiv.org/abs/2004.06376
AUTHORS:        Jamie Watson ; Michael Firman ; Aron Monszpart ; Gabriel J. Brostow
COMMENTS:       Accepted to CVPR 2020 as an oral presentation
HIGHLIGHT:      We introduce a model to predict the geometry of both visible and occluded traversable surfaces, given a single RGB image as input.

30, TITLE:      Exact MAP-Inference by Confining Combinatorial Search with LP Relaxation
http://arxiv.org/abs/2004.06370
AUTHORS:        Stefan Haller ; Paul Swoboda ; Bogdan Savchynskyy
COMMENTS:       32nd AAAI Conference on Artificial Intelligence, 2018
HIGHLIGHT:      We propose a family of relaxations (different from the famous Sherali-Adams hierarchy), which naturally define lower bounds for its optimum.

31, TITLE:      SpeedNet: Learning the Speediness in Videos
http://arxiv.org/abs/2004.06130
AUTHORS:        Sagie Benaim ; Ariel Ephrat ; Oran Lang ; Inbar Mosseri ; William T. Freeman ; Michael Rubinstein ; Michal Irani ; Tali Dekel
COMMENTS:       Accepted to CVPR 2020 (oral). Project webpage: http://speednet-cvpr20.github.io
HIGHLIGHT:      We show how this single, binary classification network can be used to detect arbitrary rates of speediness of objects.

32, TITLE:      Simple Multi-Resolution Representation Learning for Human Pose Estimation
http://arxiv.org/abs/2004.06366
AUTHORS:        Trung Q. Tran ; Giang V. Nguyen ; Daeyoung Kim
HIGHLIGHT:      In this paper, we introduce novel network structures referred to as multiresolution representation learning for human keypoint prediction.

33, TITLE:      FineGym: A Hierarchical Video Dataset for Fine-grained Action Understanding
http://arxiv.org/abs/2004.06704
AUTHORS:        Dian Shao ; Yue Zhao ; Bo Dai ; Dahua Lin
COMMENTS:       CVPR 2020 Oral (3 strong accepts); Project page: https://sdolivia.github.io/FineGym/
HIGHLIGHT:      We systematically investigate representative methods on this dataset and obtain a number of interesting findings.

34, TITLE:      The $(1 + (λ, λ))$ GA Is Even Faster on Multimodal Problems
http://arxiv.org/abs/2004.06702
AUTHORS:        Denis Antipov ; Benjamin Doerr ; Vitalii Karavaev
COMMENTS:       The full version of the paper presented at GECCO 2020 containing all the proofs omitted in the conference paper
HIGHLIGHT:      In this work, we conduct the first runtime analysis of this algorithm on a multimodal problem class, the jump functions benchmark.

35, TITLE:      Deformable Siamese Attention Networks for Visual Object Tracking
http://arxiv.org/abs/2004.06711
AUTHORS:        Yuechen Yu ; Yilei Xiong ; Weilin Huang ; Matthew R. Scott
COMMENTS:       To appear in CVPR 2020
HIGHLIGHT:      In this paper, we propose Deformable Siamese Attention Networks, referred to as SiamAttn, by introducing a new Siamese attention mechanism that computes deformable self-attention and cross-attention.

36, TITLE:      Unsupervised Multimodal Video-to-Video Translation via Self-Supervised Learning
http://arxiv.org/abs/2004.06502
AUTHORS:        Kangning Liu ; Shuhang Gu ; Andres Romero ; Radu Timofte
HIGHLIGHT:      In this work, we propose UVIT, a novel unsupervised video-to-video translation model.

37, TITLE:      SpaceNet 6: Multi-Sensor All Weather Mapping Dataset
http://arxiv.org/abs/2004.06500
AUTHORS:        Jacob Shermeyer ; Daniel Hogan ; Jason Brown ; Adam Van Etten ; Nicholas Weir ; Fabio Pacifici ; Ronny Haensch ; Alexei Bastidas ; Scott Soenen ; Todd Bacastow ; Ryan Lewis
COMMENTS:       To appear in CVPR EarthVision Proceedings, 10 pages, 7 figures
HIGHLIGHT:      To address this problem, we present an open Multi-Sensor All Weather Mapping (MSAW) dataset and challenge, which features two collection modalities (both SAR and optical). We present a baseline and benchmark for building footprint extraction with SAR data and find that state-of-the-art segmentation models pre-trained on optical data, and then trained on SAR (F1 score of 0.21) outperform those trained on SAR data alone (F1 score of 0.135).

38, TITLE:      Contrastive Examples for Addressing the Tyranny of the Majority
http://arxiv.org/abs/2004.06524
AUTHORS:        Viktoriia Sharmanska ; Lisa Anne Hendricks ; Trevor Darrell ; Novi Quadrianto
HIGHLIGHT:      Whenever a causal graph is available, we can put those contrastive examples in the perspective of counterfactuals.

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

40, TITLE:      Towards Realistic Optimization Benchmarks: A Questionnaire on the Properties of Real-World Problems
http://arxiv.org/abs/2004.06395
AUTHORS:        Koen van der Blom ; Timo M. Deist ; Tea Tušar ; Mariapia Marchi ; Yusuke Nojima ; Akira Oyama ; Vanessa Volz ; Boris Naujoks
COMMENTS:       2 pages, GECCO2020 Poster Paper
HIGHLIGHT:      This work aims to identify properties of real-world problems through a questionnaire on real-world single-, multi-, and many-objective optimization problems.

41, TITLE:      An Efficient UAV-based Artificial Intelligence Framework for Real-Time Visual Tasks
http://arxiv.org/abs/2004.06154
AUTHORS:        Enkhtogtokh Togootogtokh ; Christian Micheloni ; Gian Luca Foresti ; Niki Martinel
HIGHLIGHT:      In this paper we focus on this challenge and introduce a multi-layer AI (MLAI) framework to allow easy integration of ad-hoc visual-based AI applications.

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

43, TITLE:      Kinship Identification through Joint Learning Using Kinship Verification Ensemble
http://arxiv.org/abs/2004.06382
AUTHORS:        Wei Wang ; Shaodi You ; Sezer Karaoglu ; Theo Gevers
COMMENTS:       18 pages, 8 figures
HIGHLIGHT:      To solve it, we propose a novel kinship identification approach through the joint training of kinship verification ensembles and a Joint Identification Module.

44, TITLE:      Event detection in coarsely annotated sports videos via parallel multi receptive field 1D convolutions
http://arxiv.org/abs/2004.06172
AUTHORS:        Kanav Vats ; Mehrnaz Fani ; Pascale Walters ; David A. Clausi ; John Zelek
HIGHLIGHT:      We introduce a multi-tower temporal convolutional network architecture for the proposed task.

45, TITLE:      AREDSUM: Adaptive Redundancy-Aware Iterative Sentence Ranking for Extractive Document Summarization
http://arxiv.org/abs/2004.06176
AUTHORS:        Keping Bi ; Rahul Jha ; W. Bruce Croft ; Asli Celikyilmaz
HIGHLIGHT:      Building on the state-of-the-art encoding methods for summarization, we present two adaptive learning models: AREDSUM-SEQ that jointly considers salience and novelty during sentence selection; and a two-step AREDSUM-CTX that scores salience first, then learns to balance salience and redundancy, enabling the measurement of the impact of each aspect.

46, TITLE:      Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks
http://arxiv.org/abs/2004.06165
AUTHORS:        Xiujun Li ; Xi Yin ; Chunyuan Li ; Xiaowei Hu ; Pengchuan Zhang ; Lei Zhang ; Lijuan Wang ; Houdong Hu ; Li Dong ; Furu Wei ; Yejin Choi ; Jianfeng Gao
HIGHLIGHT:      Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks

47, TITLE:      Blind Quality Assessment for Image Superresolution Using Deep Two-Stream Convolutional Networks
http://arxiv.org/abs/2004.06163
AUTHORS:        Wei Zhou ; Qiuping Jiang ; Yuwang Wang ; Zhibo Chen ; Weiping Li
HIGHLIGHT:      In this paper, we propose a no-reference/blind deep neural network-based SR image quality assessor (DeepSRQ).

48, TITLE:      A Divide-and-Conquer Approach to the Summarization of Academic Articles
http://arxiv.org/abs/2004.06190
AUTHORS:        Alexios Gidiotis ; Grigorios Tsoumakas
COMMENTS:       submitted to Information Processing and Management journal
HIGHLIGHT:      We present a novel divide-and-conquer method for the summarization of long documents.

49, TITLE:      Relation Transformer Network
http://arxiv.org/abs/2004.06193
AUTHORS:        Rajat Koner ; Poulami Sinhamahapatra ; Volker Tresp
HIGHLIGHT:      In this work, we present the Relation Transformer Network, which is a customized transformer-based architecture that models complex object to object and edge to object interactions, by taking into account global context.

50, TITLE:      PoKi: A Large Dataset of Poems by Children
http://arxiv.org/abs/2004.06188
AUTHORS:        Will E. Hipson ; Saif M. Mohammad
HIGHLIGHT:      We present a new corpus of child-written text, PoKi, which includes about 62 thousand poems written by children from grades 1 to 12.

51, TITLE:      Challenges and Opportunities for Computer Vision in Real-life Soccer Analytics
http://arxiv.org/abs/2004.06180
AUTHORS:        Neha Bhargava ; Fabio Cuzzolin
HIGHLIGHT:      In this paper, we explore some of the applications of computer vision to sports analytics.

52, TITLE:      Jointly Modeling Aspect and Sentiment with Dynamic Heterogeneous Graph Neural Networks
http://arxiv.org/abs/2004.06427
AUTHORS:        Shu Liu ; Wei Li ; Yunfang Wu ; Qi Su ; Xu Sun
HIGHLIGHT:      In this paper, we propose a novel dynamic heterogeneous graph to jointly model the two objectives in an explicit way.

53, TITLE:      Exploring Cell counting with Neural Arithmetic Logic Units
http://arxiv.org/abs/2004.06674
AUTHORS:        Ashish Rana ; Taranveer Singh ; Harpreet Singh ; Neeraj Kumar ; Prashant Singh Rana
COMMENTS:       Referenced code repository gets an update upon paper acceptance
HIGHLIGHT:      By making better predictions for higher ranges of cell count we are aiming to create better generalization systems for cell counting.

54, TITLE:      Rapid Damage Assessment Using Social Media Images by Combining Human and Machine Intelligence
http://arxiv.org/abs/2004.06675
AUTHORS:        Muhammad Imran ; Firoj Alam ; Umair Qazi ; Steve Peterson ; Ferda Ofli
COMMENTS:       Accepted at ISCRAM 2020 conference
HIGHLIGHT:      This work analyzes the usefulness of social media imagery content to perform rapid damage assessment during a real-world disaster.

55, TITLE:      An automatic COVID-19 CT segmentation based on U-Net with attention mechanism
http://arxiv.org/abs/2004.06673
AUTHORS:        Tongxue Zhou ; Stéphane Canu ; Su Ruan
HIGHLIGHT:      In this paper, we propose a U-Net based segmentation network using attention mechanism.

56, TITLE:      Weight Poisoning Attacks on Pre-trained Models
http://arxiv.org/abs/2004.06660
AUTHORS:        Keita Kurita ; Paul Michel ; Graham Neubig
COMMENTS:       Published as a long paper at ACL 2020
HIGHLIGHT:      In this paper, we show that it is possible to construct ``weight poisoning'' attacks where pre-trained weights are injected with vulnerabilities that expose ``backdoors'' after fine-tuning, enabling the attacker to manipulate the model prediction simply by injecting an arbitrary keyword.

57, TITLE:      Embedded Large-Scale Handwritten Chinese Character Recognition
http://arxiv.org/abs/2004.06209
AUTHORS:        Youssouf Chherawala ; Hans J. G. A. Dolfing ; Ryan S. Dixon ; Jerome R. Bellegarda
COMMENTS:       5 pages, 7 figures
HIGHLIGHT:      This paper describes how the Apple deep learning recognition system can accurately handle up to 30,000 Chinese characters while running in real-time across a range of mobile devices.

58, TITLE:      Combined Model for Partially-Observable and Non-Observable Task Switching: Solving Hierarchical Reinforcement Learning Problems Statically and Dynamically with Transfer Learning
http://arxiv.org/abs/2004.06213
AUTHORS:        Nibraas Khan ; Joshua Phillips
COMMENTS:       8 pages. arXiv admin note: substantial text overlap with arXiv:1911.10425
HIGHLIGHT:      We propose a new model, PONOWMtk, which combines both approaches, ATRs and input storage, with a static or dynamic number of ATRs.

59, TITLE:      Tensor Network Rewriting Strategies for Satisfiability and Counting
http://arxiv.org/abs/2004.06455
AUTHORS:        Niel de Beaudrap ; Aleks Kissinger ; Konstantinos Meichanetzidis
COMMENTS:       12 pages, submitted to QPL 2020
HIGHLIGHT:      We provide a graphical treatment of SAT and \#SAT on equal footing.

60, TITLE:      Robustly Pre-trained Neural Model for Direct Temporal Relation Extraction
http://arxiv.org/abs/2004.06216
AUTHORS:        Hong Guan ; Jianfu Li ; Hua Xu ; Murthy Devarakonda
COMMENTS:       10 pages, 1 Figure, 7 Tables
HIGHLIGHT:      Methods: We studied several variants of BERT (Bidirectional Encoder Representations using Transformers) some involving clinical domain customization and the others involving improved architecture and/or training strategies.

61, TITLE:      DialGraph: Sparse Graph Learning Networks for Visual Dialog
http://arxiv.org/abs/2004.06698
AUTHORS:        Gi-Cheon Kang ; Junseok Park ; Hwaran Lee ; Byoung-Tak Zhang ; Jin-Hwa Kim
COMMENTS:       14 pages, 5 figures
HIGHLIGHT:      In this paper, we formulate the visual dialog tasks as graph structure learning tasks.

62, TITLE:      Query-Variant Advertisement Text Generation with Association Knowledge
http://arxiv.org/abs/2004.06438
AUTHORS:        Siyu Duan ; Wei Li ; Cai Jing ; Yancheng He ; Yunfang Wu ; Xu Sun
HIGHLIGHT:      In this paper, we propose the query-variant advertisement text generation task that aims to generate candidate advertisements for different queries with various needs given the item keywords.

63, TITLE:      The quantum query complexity of composition with a relation
http://arxiv.org/abs/2004.06439
AUTHORS:        Aleksandrs Belovs ; Troy Lee
COMMENTS:       15 pages
HIGHLIGHT:      In this note we show a perfect composition theorem for the composition of a relation $f$ with a Boolean function $g$ \[ \mathrm{ADV}_{rel}^\pm(f \circ g^n) = \mathrm{ADV}_{rel}^\pm(f) \mathrm{ADV}^\pm(g) \enspace .

64, TITLE:      Reverse Engineering Configurations of Neural Text Generation Models
http://arxiv.org/abs/2004.06201
AUTHORS:        Yi Tay ; Dara Bahri ; Che Zheng ; Clifford Brunk ; Donald Metzler ; Andrew Tomkins
COMMENTS:       ACL 2020
HIGHLIGHT:      In the spirit of better understanding generative text models and their artifacts, we propose the new task of distinguishing which of several variants of a given model generated a piece of text, and we conduct an extensive suite of diagnostic tests to observe whether modeling choices (e.g., sampling methods, top-$k$ probabilities, model architectures, etc.) leave detectable artifacts in the text they generate.

65, TITLE:      Multi-Resolution A*
http://arxiv.org/abs/2004.06684
AUTHORS:        Wei Du ; Fahad Islam ; Maxim Likhachev
HIGHLIGHT:      To effectively leverage the advantages of both high and low resolution discretizations, we propose Multi-Resolution A* (MRA*) algorithm, that runs multiple weighted-A*(WA*) searches having different resolution levels simultaneously and combines the strengths of all of them.

66, TITLE:      Weakly Supervised Deep Learning for COVID-19 Infection Detection and Classification from CT Images
http://arxiv.org/abs/2004.06689
AUTHORS:        Shaoping Hu ; Yuan Gao ; Zhangming Niu ; Yinghui Jiang ; Lao Li ; Xianglu Xiao ; Minhao Wang ; Evandro Fei Fang ; Wade Menpes-Smith ; Jun Xia ; Hui Ye ; Guang Yang
COMMENTS:       21 pages, 7 figures
HIGHLIGHT:      In this study, we propose a weakly supervised deep learning strategy for detecting and classifying COVID-19 infection from CT images.

67, TITLE:      End-to-End Variational Networks for Accelerated MRI Reconstruction
http://arxiv.org/abs/2004.06688
AUTHORS:        Anuroop Sriram ; Jure Zbontar ; Tullie Murrell ; Aaron Defazio ; C. Lawrence Zitnick ; Nafissa Yakubova ; Florian Knoll ; Patricia Johnson
HIGHLIGHT:      In this paper, we present a new approach to this problem that extends previously proposed variational methods by learning fully end-to-end.

68, TITLE:      Imitation Learning for Fashion Style Based on Hierarchical Multimodal Representation
http://arxiv.org/abs/2004.06229
AUTHORS:        Shizhu Liu ; Shanglin Yang ; Hui Zhou
HIGHLIGHT:      In this work, we propose an adversarial inverse reinforcement learning formulation to recover reward functions based on hierarchical multimodal representation (HM-AIRL) during the imitation process.

69, TITLE:      End-to-End Estimation of Multi-Person 3D Poses from Multiple Cameras
http://arxiv.org/abs/2004.06239
AUTHORS:        Hanyue Tu ; Chunyu Wang ; Wenjun Zeng
HIGHLIGHT:      We present an approach to estimate 3D poses of multiple people from multiple camera views.

70, TITLE:      A Robust Reputation-based Group Ranking System and its Resistance to Bribery
http://arxiv.org/abs/2004.06223
AUTHORS:        Joao Saude ; Guilherme Ramos ; Ludovico Boratto
COMMENTS:       28 pages, 14 figures
HIGHLIGHT:      In this paper, we propose a new reputation-based ranking system, utilizing multipartite rating subnetworks, which clusters users by their similarities using three measures, two of them based on Kolmogorov complexity.

71, TITLE:      Deep Learning Models for Multilingual Hate Speech Detection
http://arxiv.org/abs/2004.06465
AUTHORS:        Sai Saket Aluru ; Binny Mathew ; Punyajoy Saha ; Animesh Mukherjee
COMMENTS:       16 pages
HIGHLIGHT:      In this paper, we conduct a large scale analysis of multilingual hate speech in 9 languages from 16 different sources.

72, TITLE:      Cascade Neural Ensemble for Identifying Scientifically Sound Articles
http://arxiv.org/abs/2004.06222
AUTHORS:        Ashwin Karthik Ambalavanan ; Murthy Devarakonda
COMMENTS:       11 pages, 4 figures, and 9 tables
HIGHLIGHT:      Since scientifically sound articles are identified through a multi-step process we proposed a novel cascade ensemble analogous to the selection process.

73, TITLE:      Self6D: Self-Supervised Monocular 6D Object Pose Estimation
http://arxiv.org/abs/2004.06468
AUTHORS:        Gu Wang ; Fabian Manhardt ; Jianzhun Shao ; Xiangyang Ji ; Nassir Navab ; Federico Tombari
HIGHLIGHT:      To overcome this shortcoming, we propose the idea of monocular 6D pose estimation by means of self-supervised learning, which eradicates the need for real data with annotations.

74, TITLE:      What's so special about BERT's layers? A closer look at the NLP pipeline in monolingual and multilingual models
http://arxiv.org/abs/2004.06499
AUTHORS:        Wietse de Vries ; Andreas van Cranenburgh ; Malvina Nissim
HIGHLIGHT:      We investigate to what extent these results also hold for a language other than English.

75, TITLE:      PhICNet: Physics-Incorporated Convolutional Recurrent Neural Networks for Modeling Dynamical Systems
http://arxiv.org/abs/2004.06243
AUTHORS:        Priyabrata Saha ; Saurabh Dash ; Saibal Mukhopadhyay
HIGHLIGHT:      In this paper, we present a physics-incorporated deep learning framework to model and predict the spatiotemporal evolution of dynamical systems governed by partially-known inhomogenous PDEs with unobservable source dynamics.

76, TITLE:      A reinforcement learning application of guided Monte Carlo Tree Search algorithm for beam orientation selection in radiation therapy
http://arxiv.org/abs/2004.06244
AUTHORS:        Azar Sadeghnejad-Barkousaraie ; Gyanendra Bohara ; Steve Jiang ; Dan Nguyen
HIGHLIGHT:      We propose a reinforcement learning strategy using Monte Carlo Tree Search capable of finding a superior beam orientation set and in less time than CG.We utilized a reinforcement learning structure involving a supervised learning network to guide Monte Carlo tree search (GTS) to explore the decision space of beam orientation selection problem.

77, TITLE:      Multi-source Attention for Unsupervised Domain Adaptation
http://arxiv.org/abs/2004.06608
AUTHORS:        Xia Cui ; Danushka Bollegala
HIGHLIGHT:      Domain adaptation considers the problem of generalising a model learnt using data from a particular source domain to a different target domain.

78, TITLE:      Dichotomy for Graph Homomorphisms with Complex Values on Bounded Degree Graphs
http://arxiv.org/abs/2004.06620
AUTHORS:        Jin-Yi Cai ; Artem Govorov
HIGHLIGHT:      The complexity of graph homomorphisms has been a subject of intense study [11, 12, 4, 42, 21, 17, 6, 20].

79, TITLE:      Divergence-Based Adaptive Extreme Video Completion
http://arxiv.org/abs/2004.06409
AUTHORS:        Majed El Helou ; Ruofan Zhou ; Frank Schmutz ; Fabrice Guibert ; Sabine Süsstrunk
HIGHLIGHT:      We propose an extension of a state-of-the-art extreme image completion algorithm to extreme video completion.

80, TITLE:      A Transfer Learning approach to Heatmap Regression for Action Unit intensity estimation
http://arxiv.org/abs/2004.06657
AUTHORS:        Ioanna Ntinou ; Enrique Sanchez ; Adrian Bulat ; Michel Valstar ; Georgios Tzimiropoulos
COMMENTS:       Submitted for review to IEEE Trans. on Affective Computing
HIGHLIGHT:      To this end, we propose a simple yet efficient approach based on Heatmap Regression that merges both problems into a single task. Motivated by this observation we propose a novel AU modelling problem that consists of jointly estimating their localisation and intensity.

81, TITLE:      Distilling Localization for Self-Supervised Representation Learning
http://arxiv.org/abs/2004.06638
AUTHORS:        Nanxuan Zhao ; Zhirong Wu ; Rynson W. H. Lau ; Stephen Lin
HIGHLIGHT:      To address this problem, we propose a data-driven approach for learning invariance to backgrounds.

82, TITLE:      An Attention-Based System for Damage Assessment Using Satellite Imagery
http://arxiv.org/abs/2004.06643
AUTHORS:        Hanxiang Hao ; Sriram Baireddy ; Emily R. Bartusiak ; Latisha Konz ; Kevin LaTourette ; Michael Gribbons ; Moses Chan ; Mary L. Comer ; Edward J. Delp
COMMENTS:       10 pages, 9 figures
HIGHLIGHT:      In this paper, we present Siam-U-Net-Attn model - a multi-class deep learning model with an attention mechanism - to assess damage levels of buildings given a pair of satellite images depicting a scene before and after a disaster.

83, TITLE:      StandardGAN: Multi-source Domain Adaptation for Semantic Segmentation of Very High Resolution Satellite Images by Data Standardization
http://arxiv.org/abs/2004.06402
AUTHORS:        Onur Tasar ; Yuliya Tarabalka ; Alain Giros ; Pierre Alliez ; Sébastien Clerc
COMMENTS:       Accepted at CVPR EarthVision Workshop 2020
HIGHLIGHT:      In this work, we deal with the multi-source domain adaptation problem.



==========Updates to Previous Papers==========
1, TITLE:       From Data to Actions in Intelligent Transportation Systems: a Prescription of Functional Requirements for Model Actionability
http://arxiv.org/abs/2002.02210
AUTHORS:        Ibai Lana ; Javier J. Sanchez-Medina ; Eleni I. Vlahogianni ; Javier Del Ser
COMMENTS:       22 pages, 3 figures
HIGHLIGHT:      This work aims to describe how data, coming from diverse ITS sources, can be used to learn and adapt data-driven models for efficiently operating ITS assets, systems and processes; in other words, for data-based models to fully become actionable.

2, TITLE:       Person Re-Identification via Active Hard Sample Mining
http://arxiv.org/abs/2004.04912
AUTHORS:        Xin Xu ; Lei Liu ; Weifeng Liu ; Meng Wang ; Ruimin Hu
HIGHLIGHT:      To alleviate such a problem, we present an active hard sample mining framework via training an effective re-ID model with the least labeling efforts.

3, TITLE:       SSRNet: Scalable 3D Surface Reconstruction Network
http://arxiv.org/abs/1911.07401
AUTHORS:        Zhenxing Mi ; Yiming Luo ; Wenbing Tao
COMMENTS:       Accepted by CVPR2020, typos corrected, references added, images revised
HIGHLIGHT:      In this paper, we propose the SSRNet, a novel scalable learning-based method for surface reconstruction.

4, TITLE:       3D Photography using Context-aware Layered Depth Inpainting
http://arxiv.org/abs/2004.04727
AUTHORS:        Meng-Li Shih ; Shih-Yang Su ; Johannes Kopf ; Jia-Bin Huang
COMMENTS:       CVPR 2020. Project page: https://shihmengli.github.io/3D-Photo-Inpainting/ Code: https://github.com/vt-vl-lab/3d-photo-inpainting Demo: https://colab.research.google.com/drive/1706ToQrkIZshRSJSHvZ1RuCiM__YX3Bz
HIGHLIGHT:      We propose a method for converting a single RGB-D input image into a 3D photo - a multi-layer representation for novel view synthesis that contains hallucinated color and depth structures in regions occluded in the original view.

5, TITLE:       Equalization Loss for Long-Tailed Object Recognition
http://arxiv.org/abs/2003.05176
AUTHORS:        Jingru Tan ; Changbao Wang ; Buyu Li ; Quanquan Li ; Wanli Ouyang ; Changqing Yin ; Junjie Yan
COMMENTS:       CVPR 2020. Winner of LVIS Challenge 2019. Code has been available at https: //github.com/tztztztztz/eql.detectron2
HIGHLIGHT:      In this work, we analyze this problem from a novel perspective: each positive sample of one category can be seen as a negative sample for other categories, making the tail categories receive more discouraging gradients.

6, TITLE:       Re-translation versus Streaming for Simultaneous Translation
http://arxiv.org/abs/2004.03643
AUTHORS:        Naveen Arivazhagan ; Colin Cherry ; Wolfgang Macherey ; George Foster
HIGHLIGHT:      We study a related problem in which revisions to the hypothesis beyond strictly appending words are permitted.

7, TITLE:       Passage Re-ranking with BERT
http://arxiv.org/abs/1901.04085
AUTHORS:        Rodrigo Nogueira ; Kyunghyun Cho
HIGHLIGHT:      In this paper, we describe a simple re-implementation of BERT for query-based passage re-ranking.

8, TITLE:       The relationship between Fully Connected Layers and number of classes for the analysis of retinal images
http://arxiv.org/abs/2004.03624
AUTHORS:        Ajna Ram ; Constantino Carlos Reyes-Aldasoro
HIGHLIGHT:      This paper hence aims to find the relationship between number of classes and number of fully-connected layers.

9, TITLE:       Translation Artifacts in Cross-lingual Transfer Learning
http://arxiv.org/abs/2004.04721
AUTHORS:        Mikel Artetxe ; Gorka Labaka ; Eneko Agirre
HIGHLIGHT:      In this paper, we show that such translation process can introduce subtle artifacts that have a notable impact in existing cross-lingual models.

10, TITLE:      Enriching Consumer Health Vocabulary Using Enhanced GloVe Word Embedding
http://arxiv.org/abs/2004.00150
AUTHORS:        Mohammed Ibrahim ; Susan Gauch ; Omar Salman ; Mohammed Alqahatani
HIGHLIGHT:      In this paper, we present an enhanced word embedding technique that generates new CHV terms from a consumer-generated text.

11, TITLE:      Multiagent Rollout Algorithms and Reinforcement Learning
http://arxiv.org/abs/1910.00120
AUTHORS:        Dimitri Bertsekas
HIGHLIGHT:      We introduce an approach, whereby at every stage, each agent's decision is made by executing a local rollout algorithm that uses a base policy, together with some coordinating information from the other agents.

12, TITLE:      Adversarial Examples Improve Image Recognition
http://arxiv.org/abs/1911.09665
AUTHORS:        Cihang Xie ; Mingxing Tan ; Boqing Gong ; Jiang Wang ; Alan Yuille ; Quoc V. Le
COMMENTS:       CVPR 2020, models are available at https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet
HIGHLIGHT:      Here we present an opposite perspective: adversarial examples can be used to improve image recognition models if harnessed in the right manner.

13, TITLE:      Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder
http://arxiv.org/abs/2003.02977
AUTHORS:        Zhisheng Xiao ; Qing Yan ; Yali Amit
HIGHLIGHT:      In this paper, we make the observation that some of these methods fail when applied to generative models based on Variational Auto-encoders (VAE).

14, TITLE:      Meta Reinforcement Learning with Autonomous Inference of Subtask Dependencies
http://arxiv.org/abs/2001.00248
AUTHORS:        Sungryull Sohn ; Hyunjae Woo ; Jongwook Choi ; Honglak Lee
COMMENTS:       Published in ICLR 2020
HIGHLIGHT:      We propose and address a novel few-shot RL problem, where a task is characterized by a subtask graph which describes a set of subtasks and their dependencies that are unknown to the agent.

15, 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".

16, TITLE:      MGGR: MultiModal-Guided Gaze Redirection with Coarse-to-Fine Learning
http://arxiv.org/abs/2004.03064
AUTHORS:        Jingjing Chen ; Jichao Zhang ; Jiayuan Fan ; Tao Chen ; Enver Sangineto ; Nicu Sebe
HIGHLIGHT:      To this end, we propose an innovative MultiModal-Guided Gaze Redirection~(MGGR) framework that fully exploits eye-map images and target angles to adjust a given eye appearance through a designed coarse-to-fine learning.

17, TITLE:      ModuleNet: Knowledge-inherited Neural Architecture Search
http://arxiv.org/abs/2004.05020
AUTHORS:        Yaran Chen ; Ruiyuan Gao ; Fenggang Liu ; Dongbin Zhao
HIGHLIGHT:      In this paper, we discuss what kind of knowledge in a model can and should be used for new architecture design.

18, TITLE:      PuckNet: Estimating hockey puck location from broadcast video
http://arxiv.org/abs/1912.05107
AUTHORS:        Kanav Vats ; William McNally ; Chris Dulhanty ; Zhong Qiu Lin ; David A. Clausi ; John Zelek
COMMENTS:       Accepted to the Artificial Intelligence in Team Sports Workshop in AAAI 2020
HIGHLIGHT:      We introduce a novel methodology for determining puck location from approximate puck location annotations in broadcast video.

19, TITLE:      A Story of Two Streams: Reinforcement Learning Models from Human Behavior and Neuropsychiatry
http://arxiv.org/abs/1906.11286
AUTHORS:        Baihan Lin ; Guillermo Cecchi ; Djallel Bouneffouf ; Jenna Reinen ; Irina Rish
COMMENTS:       Published in AAMAS 2020 as a full paper. This article supersedes our work arXiv:1706.02897 into RL setting and extends extensively into RL games, cognitive modeling, and gambling tasks in lifelong learning setting
HIGHLIGHT:      Drawing an inspiration from behavioral studies of human decision making, we propose here a more general and flexible parametric framework for reinforcement learning that extends standard Q-learning to a two-stream model for processing positive and negative rewards, and allows to incorporate a wide range of reward-processing biases -- an important component of human decision making which can help us better understand a wide spectrum of multi-agent interactions in complex real-world socioeconomic systems, as well as various neuropsychiatric conditions associated with disruptions in normal reward processing.

20, TITLE:      An Interpretable Compression and Classification System: Theory and Applications
http://arxiv.org/abs/1907.08952
AUTHORS:        Tzu-Wei Tseng ; Kai-Jiun Yang ; C. -C. Jay Kuo ;  Shang-Ho ;  Tsai
COMMENTS:       12 pages, 12 figures and 5 tables
HIGHLIGHT:      This study proposes a low-complexity interpretable classification system.

21, TITLE:      Disentangling Human Dynamics for Pedestrian Locomotion Forecasting with Noisy Supervision
http://arxiv.org/abs/1911.01138
AUTHORS:        Karttikeya Mangalam ; Ehsan Adeli ; Kuan-Hui Lee ; Adrien Gaidon ; Juan Carlos Niebles
COMMENTS:       Accepted to WACV 2020 (Oral)
HIGHLIGHT:      We present a method to disentangle the overall pedestrian motion into easier to learn subparts by utilizing a pose completion and a decomposition module.

22, TITLE:      Just Go with the Flow: Self-Supervised Scene Flow Estimation
http://arxiv.org/abs/1912.00497
AUTHORS:        Himangi Mittal ; Brian Okorn ; David Held
COMMENTS:       Accepted at CVPR 2020 (Oral)
HIGHLIGHT:      As an alternative, we present a method of training scene flow that uses two self-supervised losses, based on nearest neighbors and cycle consistency.

23, TITLE:      Effect of Annotation Errors on Drone Detection with YOLOv3
http://arxiv.org/abs/2004.01059
AUTHORS:        Aybora Koksal ; Kutalmis Gokalp Ince ; A. Aydin Alatan
HIGHLIGHT:      In this work, different types of annotation errors for object detection problem are simulated and the performance of a popular state-of-the-art object detector, YOLOv3, with erroneous annotations during training and testing stages is examined.

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

25, TITLE:      Cumulo: A Dataset for Learning Cloud Classes
http://arxiv.org/abs/1911.04227
AUTHORS:        Valentina Zantedeschi ; Fabrizio Falasca ; Alyson Douglas ; Richard Strange ; Matt J. Kusner ; Duncan Watson-Parris
HIGHLIGHT:      In this paper, we introduce Cumulo, a benchmark dataset for training and evaluating global cloud classification models. To compare methods, we introduce a set of evaluation criteria, to identify models that are not only accurate, but also physically-realistic.

26, TITLE:      Populations of Spiking Neurons for Reservoir Computing: Closed Loop Control of a Compliant Quadruped
http://arxiv.org/abs/2004.04560
AUTHORS:        Alexander Vandesompele ; Gabriel Urbain ; Francis wyffels ; Joni Dambre
HIGHLIGHT:      Here, we present a novel framework for implementing central pattern generators with spiking neural networks to obtain closed loop robot control.

27, TITLE:      Characterizing the dynamics of learning in repeated reference games
http://arxiv.org/abs/1912.07199
AUTHORS:        Robert D. Hawkins ; Michael C. Frank ; Noah D. Goodman
COMMENTS:       Accepted at Cognitive Science
HIGHLIGHT:      The language we use over the course of conversation changes as we establish common ground and learn what our partner finds meaningful. We release an open corpus (>15,000 utterances) of extended dyadic interactions in a classic repeated reference game task where pairs of participants had to coordinate on how to refer to initially difficult-to-describe tangram stimuli.

28, TITLE:      Safe Execution of Concurrent Programs by Enforcement of Scheduling Constraints
http://arxiv.org/abs/1809.01955
AUTHORS:        Patrick Metzler ; Habib Saissi ; Péter Bokor ; Neeraj Suri
HIGHLIGHT:      As strict enforcement of scheduling constraints may induce a high execution time overhead, we present optimizations over a naive solution that reduce this overhead.

29, TITLE:      D3S -- A Discriminative Single Shot Segmentation Tracker
http://arxiv.org/abs/1911.08862
AUTHORS:        Alan Lukežič ; Jiří Matas ; Matej Kristan
COMMENTS:       The paper is accepted to the CVPR2020
HIGHLIGHT:      We propose a discriminative single-shot segmentation tracker - D3S, which narrows the gap between visual object tracking and video object segmentation.

30, TITLE:      Real Time Detection of Small Objects
http://arxiv.org/abs/2003.07442
AUTHORS:        Al-Akhir Nayan ; Joyeta Saha ; Ahamad Nokib Mozumder
COMMENTS:       7 pages, 8 figures
HIGHLIGHT:      We have introduced a novel real time detection algorithm which employs upsampling and skip connection to extract multiscale features at different convolution levels in a learning task resulting a remarkable performance in detecting small objects.

31, TITLE:      Machine Translation with Cross-lingual Word Embeddings
http://arxiv.org/abs/1912.10167
AUTHORS:        Marco Berlot ; Evan Kaplan
HIGHLIGHT:      Machine Translation with Cross-lingual Word Embeddings

32, TITLE:      Generating Representative Headlines for News Stories
http://arxiv.org/abs/2001.09386
AUTHORS:        Xiaotao Gu ; Yuning Mao ; Jiawei Han ; Jialu Liu ; Hongkun Yu ; You Wu ; Cong Yu ; Daniel Finnie ; Jiaqi Zhai ; Nicholas Zukoski
COMMENTS:       WebConf 2020 (WWW 2020)
HIGHLIGHT:      In this work, we study the problem of generating representative headlines for news stories.

33, TITLE:      Learning a Generic Adaptive Wavelet Shrinkage Function for Denoising
http://arxiv.org/abs/1910.09234
AUTHORS:        Tobias Alt ; Joachim Weickert
HIGHLIGHT:      To reduce this gap, we introduce a generic wavelet shrinkage function for denoising which is adaptive to both the wavelet scales as well as the noise standard deviation.

34, TITLE:      Medical Image Segmentation via Unsupervised Convolutional Neural Network
http://arxiv.org/abs/2001.10155
AUTHORS:        Junyu Chen ; Eric C. Frey
HIGHLIGHT:      In this paper, we present a novel learning-based segmentation model that could be trained semi- or un- supervised.

35, TITLE:      Sequential Adaptive Design for Jump Regression Estimation
http://arxiv.org/abs/1904.01648
AUTHORS:        Chiwoo Park ; Peihua Qiu ; Jennifer Carpena-Núñez ; Rahul Rao ; Michael Susner ; Benji Maruyama
HIGHLIGHT:      In this paper, we develop a design strategy of selecting the design points for regression analysis with discontinuities.

36, TITLE:      Towards Visually Explaining Variational Autoencoders
http://arxiv.org/abs/1911.07389
AUTHORS:        Wenqian Liu ; Runze Li ; Meng Zheng ; Srikrishna Karanam ; Ziyan Wu ; Bir Bhanu ; Richard J. Radke ; Octavia Camps
COMMENTS:       10 pages, 9 figures, 2 tables, CVPR 2020
HIGHLIGHT:      In this work, we take a step towards bridging this crucial gap, proposing the first technique to visually explain VAEs by means of gradient-based attention.

37, TITLE:      GeneCAI: Genetic Evolution for Acquiring Compact AI
http://arxiv.org/abs/2004.04249
AUTHORS:        Mojan Javaheripi ; Mohammad Samragh ; Tara Javidi ; Farinaz Koushanfar
HIGHLIGHT:      This paper introduces GeneCAI, a novel optimization method that automatically learns how to tune per-layer compression hyper-parameters.

38, TITLE:      Finding Covid-19 from Chest X-rays using Deep Learning on a Small Dataset
http://arxiv.org/abs/2004.02060
AUTHORS:        Lawrence O. Hall ; Rahul Paul ; Dmitry B. Goldgof ; Gregory M. Goldgof
COMMENTS:       6 pages
HIGHLIGHT:      This paper explores how useful chest X-ray images can be in diagnosing COVID-19 disease.

39, TITLE:      Towards Automatic Generation of Questions from Long Answers
http://arxiv.org/abs/2004.05109
AUTHORS:        Shlok Kumar Mishra ; Pranav Goel ; Abhishek Sharma ; Abhyuday Jagannatha ; David Jacobs ; Hal Daumé III
HIGHLIGHT:      Therefore, we propose a novel evaluation benchmark to assess the performance of existing AQG systems for long-text answers.

40, TITLE:      Dual-reference Age Synthesis
http://arxiv.org/abs/1908.02671
AUTHORS:        Yuan Zhou ; Bingzhang Hu ; and Jun He ; Yu Guan ; Ling Shao
HIGHLIGHT:      In this paper, we propose a novel framework taking two images as inputs, named dual-reference age synthesis (DRAS), which approaches the task differently; instead of using "hard" age information, i.e. a fixed number, our model determines the target age in a "soft" way, by employing a second reference image.

41, TITLE:      Multi-Scale Aggregation Using Feature Pyramid Module for Text-Independent Speaker Verification
http://arxiv.org/abs/2004.03194
AUTHORS:        Youngmoon Jung ; Seong Min Kye ; Yeunju Choi ; Myunghun Jung ; Hoirin Kim
COMMENTS:       Submitted to Interspeech 2020
HIGHLIGHT:      This paper improves the MSA by using a feature pyramid module, which enhances speaker-discriminative information of features at multiple layers via a top-down pathway and lateral connections.

42, TITLE:      Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation
http://arxiv.org/abs/2003.03773
AUTHORS:        Zhedong Zheng ; Yi Yang
COMMENTS:       12 pages, 6 figures, 6 tables
HIGHLIGHT:      To overcome the problem, this paper proposes to explicitly estimate the prediction uncertainty during training to rectify the pseudo label learning for unsupervised semantic segmentation adaptation.

43, TITLE:      Hierarchical Severity Staging of Anterior Cruciate Ligament Injuries using Deep Learning with MRI Images
http://arxiv.org/abs/2003.09089
AUTHORS:        Nikan K. Namiri ; Io Flament ; Bruno Astuto ; Rutwik Shah ; Radhika Tibrewala ; Francesco Caliva ; Thomas M. Link ; Valentina Pedoia ; Sharmila Majumdar
HIGHLIGHT:      Hierarchical Severity Staging of Anterior Cruciate Ligament Injuries using Deep Learning with MRI Images

44, TITLE:      Small-Object Detection in Remote Sensing Images with End-to-End Edge-Enhanced GAN and Object Detector Network
http://arxiv.org/abs/2003.09085
AUTHORS:        Jakaria Rabbi ; Nilanjan Ray ; Matthias Schubert ; Subir Chowdhury ; Dennis Chao
COMMENTS:       This paper contains 27 pages and submitted to MDPI remote sensing journal (under review) GitHub Repository: https://github.com/Jakaria08/EESRGAN (Implementation)
HIGHLIGHT:      We propose an architecture with three components: ESRGAN, Edge Enhancement Network (EEN), and Detection network.

45, 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.

46, TITLE:      Mining Commonsense Facts from the Physical World
http://arxiv.org/abs/2002.03149
AUTHORS:        Yanyan Zou ; Wei Lu ; Xu Sun
COMMENTS:       The experiment part is not insufficient which might confuses the readers, while we are not planning to improve it for now. To ensure the quality of arxiv papers, we would like to withdraw this submission
HIGHLIGHT:      In this paper, we propose a new task of mining commonsense facts from the raw text that describes the physical world. Then we create two large annotated datasets each with approximate 200k instances for commonsense knowledge base completion.

47, TITLE:      When Does Unsupervised Machine Translation Work?
http://arxiv.org/abs/2004.05516
AUTHORS:        Kelly Marchisio ; Kevin Duh ; Philipp Koehn
COMMENTS:       Correct typo in Table 3
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.

48, TITLE:      Demographic Bias in Biometrics: A Survey on an Emerging Challenge
http://arxiv.org/abs/2003.02488
AUTHORS:        P. Drozdowski ; C. Rathgeb ; A. Dantcheva ; N. Damer ; C. Busch
COMMENTS:       15 pages, 3 figures, 3 tables. Submitted to IEEE Transactions on Technology and Society. Update after first round of peer review
HIGHLIGHT:      The main contributions of this article are: (1) an overview of the topic of algorithmic bias in the context of biometrics, (2) a comprehensive survey of the existing literature on biometric bias estimation and mitigation, (3) a discussion of the pertinent technical and social matters, and (4) an outline of the remaining challenges and future work items, both from technological and social points of view.

49, TITLE:      Plug-and-play ISTA converges with kernel denoisers
http://arxiv.org/abs/2004.03145
AUTHORS:        Ruturaj G. Gavaskar ; Kunal N. Chaudhury
COMMENTS:       5 pages, Accepted to IEEE Signal Processing Letters
HIGHLIGHT:      We prove that, under reasonable assumptions, fixed-point convergence of PnP-ISTA is indeed guaranteed for linear inverse problems such as deblurring, inpainting and superresolution (the assumptions are verifiable for inpainting).

50, TITLE:      A white-box analysis on the writer-independent dichotomy transformation applied to offline handwritten signature verification
http://arxiv.org/abs/2004.03370
AUTHORS:        Victor L. F. Souza ; Adriano L. I. Oliveira ; Rafael M. O. Cruz ; Robert Sabourin
HIGHLIGHT:      In this work, we present a white-box analysis of this approach highlighting how it handles the challenges, the dynamic selection of references through fusion function, and its application for transfer learning.