2020.05.12.txt

==========New Papers==========
1, TITLE:       Diversifying Dialogue Generation with Non-Conversational Text
http://arxiv.org/abs/2005.04346
AUTHORS:        Hui Su ; Xiaoyu Shen ; Sanqiang Zhao ; Xiao Zhou ; Pengwei Hu ; Randy Zhong ; Cheng Niu ; Jie Zhou
COMMENTS:       Accepted to ACL 2020 (long)
HIGHLIGHT:      In this paper, we propose a new perspective to diversify dialogue generation by leveraging non-conversational text. We collect a large-scale non-conversational corpus from multi sources including forum comments, idioms and book snippets.

2, TITLE:       An Investigation of Why Overparameterization Exacerbates Spurious Correlations
http://arxiv.org/abs/2005.04345
AUTHORS:        Shiori Sagawa ; Aditi Raghunathan ; Pang Wei Koh ; Percy Liang
HIGHLIGHT:      We study why overparameterization -- increasing model size well beyond the point of zero training error -- can hurt test error on minority groups despite improving average test error when there are spurious correlations in the data.

3, TITLE:       GPU Acceleration of Sparse Neural Networks
http://arxiv.org/abs/2005.04347
AUTHORS:        Aavaas Gajurel ; Sushil J. Louis ; Frederick C Harris
HIGHLIGHT:      In this paper, we use graphics processing units(GPU) to accelerate sparse and arbitrary structured neural networks.

4, TITLE:       Reference-Based Sketch Image Colorization using Augmented-Self Reference and Dense Semantic Correspondence
http://arxiv.org/abs/2005.05207
AUTHORS:        Junsoo Lee ; Eungyeup Kim ; Yunsung Lee ; Dongjun Kim ; Jaehyuk Chang ; Jaegul Choo
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      To tackle this challenge, we propose to utilize the identical image with geometric distortion as a virtual reference, which makes it possible to secure the ground truth for a colored output image.

5, TITLE:       It's Morphin' Time! Combating Linguistic Discrimination with Inflectional Perturbations
http://arxiv.org/abs/2005.04364
AUTHORS:        Samson Tan ; Shafiq Joty ; Min-Yen Kan ; Richard Socher
COMMENTS:       To appear in the Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL 2020)
HIGHLIGHT:      We perturb the inflectional morphology of words to craft plausible and semantically similar adversarial examples that expose these biases in popular NLP models, e.g., BERT and Transformer, and show that adversarially fine-tuning them for a single epoch significantly improves robustness without sacrificing performance on clean data.

6, TITLE:       Medical Image Segmentation Using a U-Net type of Architecture
http://arxiv.org/abs/2005.05218
AUTHORS:        Eshal Zahra ; Bostan Ali ; Wajahat Siddique
COMMENTS:       6 Pages
HIGHLIGHT:      More specifically, we introduce a fully supervised FC layers based pixel-wise loss at the bottleneck of the encoder branch of U-Net.

7, TITLE:       Utility-aware Privacy-preserving Data Releasing
http://arxiv.org/abs/2005.04369
AUTHORS:        Di Zhuang ; J. Morris Chang
COMMENTS:       9 pages, 2 figures, 4 tables
HIGHLIGHT:      In this work, we propose a two-step perturbation-based utility-aware privacy-preserving data releasing framework.

8, TITLE:       Semi-Supervised Dialogue Policy Learning via Stochastic Reward Estimation
http://arxiv.org/abs/2005.04379
AUTHORS:        Xinting Huang ; Jianzhong Qi ; Yu Sun ; Rui Zhang
HIGHLIGHT:      To overcome this limitation, we propose a novel reward learning approach for semi-supervised policy learning.

9, TITLE:       Chance-Constrained Trajectory Optimization for Safe Exploration and Learning of Nonlinear Systems
http://arxiv.org/abs/2005.04374
AUTHORS:        Yashwanth Kumar Nakka ; Anqi Liu ; Guanya Shi ; Anima Anandkumar ; Yisong Yue ; Soon-Jo Chung
COMMENTS:       Submitted to RA-L 2020
HIGHLIGHT:      In this paper, we present a new episodic framework to design a sub-optimal pool of motion plans that aid exploration for learning unknown residual dynamics under safety constraints.

10, TITLE:      iUNets: Fully invertible U-Nets with Learnable Up- and Downsampling
http://arxiv.org/abs/2005.05220
AUTHORS:        Christian Etmann ; Rihuan Ke ; Carola-Bibiane Schönlieb
HIGHLIGHT:      Here, we present a new fully-invertible U-Net-based architecture called the \emph{iUNet}, which allows for the application of highly memory-efficient backpropagation procedures.

11, TITLE:      AutoCLINT: The Winning Method in AutoCV Challenge 2019
http://arxiv.org/abs/2005.04373
AUTHORS:        Woonhyuk Baek ; Ildoo Kim ; Sungwoong Kim ; Sungbin Lim
COMMENTS:       9 pages, 8 figures, 4 tables
HIGHLIGHT:      In this paper, we introduce the winning method in the competition, AutoCLINT.

12, TITLE:      ICE-GAN: Identity-aware and Capsule-Enhanced GAN for Micro-Expression Recognition and Synthesis
http://arxiv.org/abs/2005.04370
AUTHORS:        Jianhui Yu ; Chaoyi Zhang ; Yang Song ; Weidong Cai
COMMENTS:       13 pages, 5 figures
HIGHLIGHT:      To this end, we propose a novel Identity-aware and Capsule-Enhanced Generative Adversarial Network (ICE-GAN), which is adversarially completed with the micro-expression synthesis (MES) task, where synthetic faces with controllable micro-expressions can be produced by the generator with distinguishable identity information to improve the MER performance.

13, TITLE:      On the Transferability of Winning Tickets in Non-Natural Image Datasets
http://arxiv.org/abs/2005.05232
AUTHORS:        Matthia Sabatelli ; Mike Kestemont ; Pierre Geurts
HIGHLIGHT:      We study the generalization properties of pruned neural networks that are the winners of the lottery ticket hypothesis on datasets of natural images.

14, TITLE:      System-Level Predictive Maintenance: Review of Research Literature and Gap Analysis
http://arxiv.org/abs/2005.05239
AUTHORS:        Kyle Miller ; Artur Dubrawski
COMMENTS:       24 pages, 3 figures
HIGHLIGHT:      This paper reviews current literature in the field of predictive maintenance from the system point of view.

15, TITLE:      Enhancing LGMD's Looming Selectivity for UAVs with Spatial-temporal Distributed Presynaptic Connection
http://arxiv.org/abs/2005.04397
AUTHORS:        Jiannan Zhao ; Hongxin Wang ; Shigang Yue
COMMENTS:       14 pages, 18 figures, 4 tables
HIGHLIGHT:      In this paper, we proposed a new LGMD model for flying robots considering distributed spatial-temporal computing for both excitation and inhibition to enhance the looming selectivity in flying scenes.

16, TITLE:      Generating Pertinent and Diversified Comments with Topic-aware Pointer-Generator Networks
http://arxiv.org/abs/2005.04396
AUTHORS:        Junheng Huang ; Lu Pan ; Kang Xu ; Weihua Peng ; Fayuan Li
HIGHLIGHT:      In this paper, we propose a novel generation model based on Topic-aware Pointer-Generator Networks (TPGN), which can utilize the topic information hidden in the articles to guide the generation of pertinent and diversified comments.

17, TITLE:      Commonsense Evidence Generation and Injection in Reading Comprehension
http://arxiv.org/abs/2005.05240
AUTHORS:        Ye Liu ; Tao Yang ; Zeyu You ; Wei Fan ; Philip S. Yu
COMMENTS:       Accepted by SIGDIAL 2020
HIGHLIGHT:      To empower the machine with commonsense reasoning, in this paper, we propose a Commonsense Evidence Generation and Injection framework in reading comprehension, named CEGI.

18, TITLE:      A Dataset for Statutory Reasoning in Tax Law Entailment and Question Answering
http://arxiv.org/abs/2005.05257
AUTHORS:        Nils Holzenberger ; Andrew Blair-Stanek ; Benjamin Van Durme
HIGHLIGHT:      To investigate the performance of natural language understanding approaches on statutory reasoning, we introduce a dataset, together with a legal-domain text corpus.

19, TITLE:      Reinforced Rewards Framework for Text Style Transfer
http://arxiv.org/abs/2005.05256
AUTHORS:        Abhilasha Sancheti ; Kundan Krishna ; Balaji Vasan Srinivasan ; Anandhavelu Natarajan
COMMENTS:       ECIR 2020
HIGHLIGHT:      In this work, we propose a reinforcement learning based framework that directly rewards the framework on these target metrics yielding a better transfer of the target style.

20, TITLE:      Toward Better Storylines with Sentence-Level Language Models
http://arxiv.org/abs/2005.05255
AUTHORS:        Daphne Ippolito ; David Grangier ; Douglas Eck ; Chris Callison-Burch
COMMENTS:       ACL 2020 short paper
HIGHLIGHT:      We propose a sentence-level language model which selects the next sentence in a story from a finite set of fluent alternatives.

21, TITLE:      Deep-Learning-based Automated Palm Tree Counting and Geolocation in Large Farms from Aerial Geotagged Images
http://arxiv.org/abs/2005.05269
AUTHORS:        Adel Ammar ; Anis Koubaa
COMMENTS:       First version of the paper, 3 pages, 2 figures
HIGHLIGHT:      In this paper, we propose a deep learning framework for the automated counting and geolocation of palm trees from aerial images using convolutional neural networks. For this purpose, we collected aerial images in a palm tree Farm in the Kharj region, in Riyadh Saudi Arabia, using DJI drones, and we built a dataset of around 10,000 instances of palms trees.

22, TITLE:      Feature Selection with Evolving, Fast and Slow Using Two Parallel Genetic Algorithms
http://arxiv.org/abs/2005.05268
AUTHORS:        Uzay Cetin ; Yunus Emre Gundogmus
HIGHLIGHT:      In this paper, we address the problem of feature selection and propose a new approach called Evolving Fast and Slow.

23, TITLE:      Fundus2Angio: A Novel Conditional GAN Architecture for Generating Fluorescein Angiography Images from Retinal Fundus Photography
http://arxiv.org/abs/2005.05267
AUTHORS:        Sharif Amit Kamran ; Khondker Fariha Hossain ; Alireza Tavakkoli ; Stewart Lee Zuckerbrod
HIGHLIGHT:      In order to eliminate the need for FA, we propose a conditional generative adversarial network (GAN)to translate fundus images to FA images.

24, TITLE:      Multidirectional Associative Optimization of Function-Specific Word Representations
http://arxiv.org/abs/2005.05264
AUTHORS:        Daniela Gerz ; Ivan Vulić ; Marek Rei ; Roi Reichart ; Anna Korhonen
COMMENTS:       ACL 2020 (Long paper)
HIGHLIGHT:      We present a neural framework for learning associations between interrelated groups of words such as the ones found in Subject-Verb-Object (SVO) structures.

25, TITLE:      A modular extension for a computer algebra system
http://arxiv.org/abs/2005.05261
AUTHORS:        Migran N. Gevorkyan ; Anna V. Korolkova ; Dmitry S. Kulyabov ; Leonid A. Sevastianov
COMMENTS:       in English; in Russian
HIGHLIGHT:      In this paper, we consider a technology for extending the SymPy computer algebra system with a low-level module that implements a random number generator.

26, TITLE:      Normalized Convolutional Neural Network
http://arxiv.org/abs/2005.05274
AUTHORS:        Dongsuk Kim ; Geonhee Lee
COMMENTS:       6pages
HIGHLIGHT:      In this paper, we propose Normalized Convolutional Neural Network(NCNN).

27, TITLE:      Using Computer Vision to enhance Safety of Workforce in Manufacturing in a Post COVID World
http://arxiv.org/abs/2005.05287
AUTHORS:        Prateek Khandelwal ; Anuj Khandelwal ; Snigdha Agarwal
COMMENTS:       6 pages, 7 figure, 1 table
HIGHLIGHT:      This paper describes an efficient and economic approach of using AI to create a safe environment in a manufacturing setup.

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

29, TITLE:      Ring Reservoir Neural Networks for Graphs
http://arxiv.org/abs/2005.05294
AUTHORS:        Claudio Gallicchio ; Alessio Micheli
COMMENTS:       Accepted for IJCNN/WCCI 2020
HIGHLIGHT:      In this paper, we study progressive simplifications to the design strategy of RC neural networks for graphs.

30, TITLE:      The Safari of Update Structures: Visiting the Lens and Quantum Enclosures
http://arxiv.org/abs/2005.05293
AUTHORS:        Matthew Wilson ; James Hefford ; Guillaume Boisseau ; Vincent Wang
HIGHLIGHT:      We build upon our recently introduced concept of an update structure to show that they are a generalisation of very-well-behaved lenses, that is, there is a bijection between a strict subset of update structures and vwb lenses in cartesian categories.

31, TITLE:      Hierarchical Regression Network for Spectral Reconstruction from RGB Images
http://arxiv.org/abs/2005.04703
AUTHORS:        Yuzhi Zhao ; Lai-Man Po ; Qiong Yan ; Wei Liu ; Tingyu Lin
COMMENTS:       1st Place in CVPRW 2020 NTIRE Spectral Reconstruction Challenge
HIGHLIGHT:      To address these problems, we propose a 4-level Hierarchical Regression Network (HRNet) with PixelShuffle layer as inter-level interaction.

32, TITLE:      CTC-synchronous Training for Monotonic Attention Model
http://arxiv.org/abs/2005.04712
AUTHORS:        Hirofumi Inaguma ; Masato Mimura ; Tatsuya Kawahara
HIGHLIGHT:      To address this problem, we propose CTC-synchronous training (CTC-ST), in which MoChA uses CTC alignments to learn optimal monotonic alignments.

33, TITLE:      Knowledge Graph semantic enhancement of input data for improving AI
http://arxiv.org/abs/2005.04726
AUTHORS:        Shreyansh Bhatt ; Amit Sheth ; Valerie Shalin ; Jinjin Zhao
HIGHLIGHT:      In this article, we discuss the use of relevant KGs to enhance input data for two applications that use machine learning -- recommendation and community detection.

34, TITLE:      Chirp Complex Cepstrum-based Decomposition for Asynchronous Glottal Analysis
http://arxiv.org/abs/2005.04724
AUTHORS:        Thomas Drugman ; Thierry Dutoit
HIGHLIGHT:      This paper proposes an extension of the complex cepstrum-based decomposition by incorporating a chirp analysis.

35, TITLE:      Dynamic IFC Theorems for Free!
http://arxiv.org/abs/2005.04722
AUTHORS:        Maximilian Algehed ; Jean-Philippe Bernardy ; Catalin Hritcu
COMMENTS:       CSF'21 summer cycle submission
HIGHLIGHT:      We show that noninterference and transparency, the key soundness theorems for dynamic IFC libraries, can be obtained "for free", as direct consequences of the more general parametricity theorem of type abstraction.

36, TITLE:      Categorical Stochastic Processes and Likelihood
http://arxiv.org/abs/2005.04735
AUTHORS:        Dan Shiebler
HIGHLIGHT:      In this work we take a Category Theoretic perspective on the relationship between probabilistic modeling and function approximation.

37, TITLE:      Structural Parameterizations of Clique Coloring
http://arxiv.org/abs/2005.04733
AUTHORS:        Lars Jaffke ; Paloma T. Lima ; Geevarghese Philip
HIGHLIGHT:      For fixed $q \ge 2$, we give an $\mathcal{O}^{\star}(q^{tw})$-time algorithm when the input graph is given together with one of its tree decompositions of width $tw$.

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

39, TITLE:      A SentiWordNet Strategy for Curriculum Learning in Sentiment Analysis
http://arxiv.org/abs/2005.04749
AUTHORS:        Vijjini Anvesh Rao ; Kaveri Anuranjana ; Radhika Mamidi
COMMENTS:       Accepted Short Paper at 25th International Conference on Applications of Natural Language to Information Systems, June 2020, DFKI Saarbr\"ucken, Germany
HIGHLIGHT:      In this paper, we apply the ideas of curriculum learning, driven by SentiWordNet in a sentiment analysis setting.

40, TITLE:      A Simple Semi-Supervised Learning Framework for Object Detection
http://arxiv.org/abs/2005.04757
AUTHORS:        Kihyuk Sohn ; Zizhao Zhang ; Chun-Liang Li ; Han Zhang ; Chen-Yu Lee ; Tomas Pfister
HIGHLIGHT:      In this paper, we propose STAC, a simple yet effective SSL framework for visual object detection along with a data augmentation strategy.

41, TITLE:      Photometric Multi-View Mesh Refinement for High-Resolution Satellite Images
http://arxiv.org/abs/2005.04777
AUTHORS:        Mathias Rothermel ; Ke Gong ; Dieter Fritsch ; Konrad Schindler ; Norbert Haala
COMMENTS:       Accepted for publication in ISPRS Journal of Photogrammetry and Remote Sensing
HIGHLIGHT:      Here, we present an approach to recover full 3D surface meshes from multi-view satellite imagery.

42, TITLE:      Knowledge Patterns
http://arxiv.org/abs/2005.04306
AUTHORS:        Peter Clark ; John Thompson ; Bruce Porter
COMMENTS:       Published in the Handbook of Ontologies, 2004, pp 191-207. (This is an updated and extended version of the paper by the same name in Proc. KR'2000)
HIGHLIGHT:      This paper describes a new technique, called "knowledge patterns", for helping construct axiom-rich, formal ontologies, based on identifying and explicitly representing recurring patterns of knowledge (theory schemata) in the ontology, and then stating how those patterns map onto domain-specific concepts in the ontology.

43, TITLE:      Measuring the Algorithmic Efficiency of Neural Networks
http://arxiv.org/abs/2005.04305
AUTHORS:        Danny Hernandez ; Tom B. Brown
COMMENTS:       20 pages, 5 figures
HIGHLIGHT:      In this work, we argue that algorithmic progress has an aspect that is both straightforward to measure and interesting: reductions over time in the compute needed to reach past capabilities.

44, TITLE:      Temporal Common Sense Acquisition with Minimal Supervision
http://arxiv.org/abs/2005.04304
AUTHORS:        Ben Zhou ; Qiang Ning ; Daniel Khashabi ; Dan Roth
COMMENTS:       Accepted by ACL 2020
HIGHLIGHT:      This work proposes a novel sequence modeling approach that exploits explicit and implicit mentions of temporal common sense, extracted from a large corpus, to build TACOLM, a temporal common sense language model.

45, TITLE:      Progressive Adversarial Semantic Segmentation
http://arxiv.org/abs/2005.04311
AUTHORS:        Abdullah-Al-Zubaer Imran ; Demetri Terzopoulos
COMMENTS:       9 pages, 5 figures, 12 tables
HIGHLIGHT:      To tackle this problem, we propose a novel end-to-end medical image segmentation model, namely Progressive Adversarial Semantic Segmentation (PASS), which can make improved segmentation predictions without requiring any domain-specific data during training time.

46, TITLE:      The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes
http://arxiv.org/abs/2005.04790
AUTHORS:        Douwe Kiela ; Hamed Firooz ; Aravind Mohan ; Vedanuj Goswami ; Amanpreet Singh ; Pratik Ringshia ; Davide Testuggine
HIGHLIGHT:      This work proposes a new challenge set for multimodal classification, focusing on detecting hate speech in multimodal memes.

47, TITLE:      ST-MNIST -- The Spiking Tactile MNIST Neuromorphic Dataset
http://arxiv.org/abs/2005.04319
AUTHORS:        Hian Hian See ; Brian Lim ; Si Li ; Haicheng Yao ; Wen Cheng ; Harold Soh ; Benjamin C. K. Tee
COMMENTS:       Corresponding authors: Benjamin C.K. Tee and Harold Soh For dataset, see http://www.benjamintee.com/stmnist 10 Pages, 4 Figures and 2 Tables
HIGHLIGHT:      Recent advancements in electronic skins have led to the development of data-driven machine learning methods that exploit this important sensory modality.

48, TITLE:      Transforming task representations to allow deep learning models to perform novel tasks
http://arxiv.org/abs/2005.04318
AUTHORS:        Andrew K. Lampinen ; James L. McClelland
COMMENTS:       34 pages
HIGHLIGHT:      To address this, we propose a general computational framework for adapting to novel tasks based on their relationship to prior tasks.

49, TITLE:      Probing Linguistic Systematicity
http://arxiv.org/abs/2005.04315
AUTHORS:        Emily Goodwin ; Koustuv Sinha ; Timothy J. O'Donnell
COMMENTS:       To appear at ACL2020, 9 pages, 2 figures
HIGHLIGHT:      Recently, there has been much interest in the question of whether deep natural language understanding models exhibit systematicity; generalizing such that units like words make consistent contributions to the meaning of the sentences in which they appear.

50, TITLE:      LinCE: A Centralized Benchmark for Linguistic Code-switching Evaluation
http://arxiv.org/abs/2005.04322
AUTHORS:        Gustavo Aguilar ; Sudipta Kar ; Thamar Solorio
COMMENTS:       Accepted to LREC 2020
HIGHLIGHT:      Recent trends in NLP research have raised an interest in linguistic code-switching (CS); modern approaches have been proposed to solve a wide range of NLP tasks on multiple language pairs. To facilitate research in this direction, we propose a centralized benchmark for Linguistic Code-switching Evaluation (LinCE) that combines ten corpora covering four different code-switched language pairs (i.e., Spanish-English, Nepali-English, Hindi-English, and Modern Standard Arabic-Egyptian Arabic) and four tasks (i.e., language identification, named entity recognition, part-of-speech tagging, and sentiment analysis).

51, TITLE:      Generalizing Outside the Training Set: When Can Neural Networks Learn Identity Effects?
http://arxiv.org/abs/2005.04330
AUTHORS:        Simone Brugiapaglia ; Matthew Liu ; Paul Tupper
HIGHLIGHT:      We provide a simple framework in which we can rigorously prove that algorithms satisfying simple criteria cannot make the correct inference.

52, TITLE:      On Weakening Strategies for PB Solvers
http://arxiv.org/abs/2005.04466
AUTHORS:        Daniel Le Berre ; Pierre Marquis ; Romain Wallon
HIGHLIGHT:      In this paper, new application strategies for this rule are studied, aiming to infer strong constraints with small coefficients.

53, TITLE:      Vehicle Re-Identification Based on Complementary Features
http://arxiv.org/abs/2005.04463
AUTHORS:        Cunyuan Gao ; Yi Hu ; Yi Zhang ; Rui Yao ; Yong Zhou ; Jiaqi Zhao
HIGHLIGHT:      In this work, we present our solution to the vehicle re-identification (vehicle Re-ID) track in AI City Challenge 2020 (AIC2020).

54, TITLE:      Visually Impaired Aid using Convolutional Neural Networks, Transfer Learning, and Particle Competition and Cooperation
http://arxiv.org/abs/2005.04473
AUTHORS:        Fabricio Breve ; Carlos Norberto Fischer
COMMENTS:       BREVE, Fabricio Aparecido; FISCHER, Carlos Norberto. Visually Impaired Aid using Convolutional Neural Networks, Transfer Learning, and Particle Competition and Cooperation In: 2020 International Joint Conference on Neural Networks (IJCNN 2020), 2020, Glasgow, UK. Proceedings of 2020 International Joint Conference on Neural Networks (IJCNN 2020), 2020. (accepted for publication)
HIGHLIGHT:      In this paper we propose the use of convolutional neural networks (CNN), transfer learning, and semi-supervised learning (SSL) to build a framework aimed at the visually impaired aid. We also propose a dataset to train the classifiers, including indoor and outdoor situations with different types of light, floor, and obstacles.

55, TITLE:      Empowering Active Learning to Jointly Optimize System and User Demands
http://arxiv.org/abs/2005.04470
AUTHORS:        Ji-Ung Lee ; Christian M. Meyer ; Iryna Gurevych
COMMENTS:       To appear as a long paper in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL 2020). Download our code and data from the user study at github: https://github.com/UKPLab/acl2020-empowering-active-learning
HIGHLIGHT:      In this paper, we propose a new active learning approach that jointly optimizes the seemingly counteracting objectives of the active learning system (training efficiently) and the user (receiving useful instances).

56, TITLE:      Generative Model-driven Structure Aligning Discriminative Embeddings for Transductive Zero-shot Learning
http://arxiv.org/abs/2005.04492
AUTHORS:        Omkar Gune ; Mainak Pal ; Preeti Mukherjee ; Biplab Banerjee ; Subhasis Chaudhuri
HIGHLIGHT:      In this work, we propose a shallow but effective neural network-based model for learning such a projection function which aligns the visual and semantic data in the latent space while simultaneously making the latent space embeddings discriminative.

57, TITLE:      Human in Events: A Large-Scale Benchmark for Human-centric Video Analysis in Complex Events
http://arxiv.org/abs/2005.04490
AUTHORS:        Weiyao Lin ; Huabin Liu ; Shizhan Liu ; Yuxi Li ; Guo-Jun Qi ; Rui Qian ; Tao Wang ; Nicu Sebe ; Ning Xu ; Hongkai Xiong ; Mubarak Shah
COMMENTS:       Dataset for ACM MM'20 Grand Challenge on Large-scale Human-centric Video Analysis in Complex Events (http://humaninevents.org ). Challenge registration deadline: May. 17. 2020
HIGHLIGHT:      To this end, we present a new large-scale dataset, named Human-in-Events or HiEve (human-centric video analysis in complex events), for understanding human motions, poses, and actions in a variety of realistic events, especially crowd & complex events.

58, TITLE:      Leveraging Monolingual Data with Self-Supervision for Multilingual Neural Machine Translation
http://arxiv.org/abs/2005.04816
AUTHORS:        Aditya Siddhant ; Ankur Bapna ; Yuan Cao ; Orhan Firat ; Mia Chen ; Sneha Kudugunta ; Naveen Arivazhagan ; Yonghui Wu
HIGHLIGHT:      In this work, we join these two lines of research and demonstrate the efficacy of monolingual data with self-supervision in multilingual NMT.

59, TITLE:      The Visual Social Distancing Problem
http://arxiv.org/abs/2005.04813
AUTHORS:        Marco Cristani ; Alessio Del Bue ; Vittorio Murino ; Francesco Setti ; Alessandro Vinciarelli
COMMENTS:       9 pages, 5 figures. All the authors equally contributed to this manuscript and they are listed by alphabetical order. Under submission
HIGHLIGHT:      To this end, we introduce the Visual Social Distancing (VSD) problem, defined as the automatic estimation of the inter-personal distance from an image, and the characterization of the related people aggregations.

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

61, TITLE:      Learning Descriptors Invariance Through Equivalence Relations Within Manifold: A New Approach to Expression Invariant 3D Face Recognition
http://arxiv.org/abs/2005.04823
AUTHORS:        Faisal R. Al-Osaimi
COMMENTS:       This paper was submitted to pattern recognition in 2015
HIGHLIGHT:      This paper presents a unique approach for the dichotomy between useful and adverse variations of key-point descriptors, namely the identity and the expression variations in the descriptor (feature) space.

62, TITLE:      Synaptic Learning with Augmented Spikes
http://arxiv.org/abs/2005.04820
AUTHORS:        Qiang Yu ; Shiming Song ; Chenxiang Ma ; Linqiang Pan ; Kay Chen Tan
COMMENTS:       13 pages
HIGHLIGHT:      In this paper, we introduce a concept of augmented spikes to carry complementary information with spike coefficients in addition to spike latencies.

63, TITLE:      Compositional Scientific Computing with Catlab and SemanticModels
http://arxiv.org/abs/2005.04831
AUTHORS:        Micah Halter ; Evan Patterson ; Andrew Baas ; James Fairbanks
COMMENTS:       3 pages, 6 figures, Applied Category Theory 2020 conference
HIGHLIGHT:      We present Catlab.jl, which provides the category-theoretic infrastructure for this project, together with SemanticModels.jl, which leverages this infrastructure for particular modeling tasks.

64, TITLE:      Non-iterative Simultaneous Rigid Registration Method for Serial Sections of Biological Tissue
http://arxiv.org/abs/2005.04848
AUTHORS:        Chang Shu ; Xi Chen ; Qiwei Xie ; Chi Xiao ; Hua Han
COMMENTS:       appears in IEEE International Symposium on Biomedical Imaging 2018 (ISBI 2018)
HIGHLIGHT:      In this paper, we propose a novel non-iterative algorithm to simultaneously estimate optimal rigid transformation for serial section images, which is a key component in volume reconstruction of serial sections of biological tissue.

65, TITLE:      Fair Division: The Computer Scientist's Perspective
http://arxiv.org/abs/2005.04855
AUTHORS:        Toby Walsh
COMMENTS:       To appear in Proceedings of IJCAI 2020
HIGHLIGHT:      I survey recent progress on a classic and challenging problem in social choice: the fair division of indivisible items.

66, TITLE:      Scope Head for Accurate Localizationin Object Detection
http://arxiv.org/abs/2005.04854
AUTHORS:        Geng Zhan ; Dan Xu ; Guo Lu ; Wei Wu ; Chunhua Shen ; Wanli Ouyang
HIGHLIGHT:      To tackle these issues, in this paper, we propose a novel detector coined as ScopeNet, which models anchors of each location as a mutually dependent relationship.

67, TITLE:      Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition
http://arxiv.org/abs/2005.04862
AUTHORS:        Ye Bai ; Jiangyan Yi ; Jianhua Tao ; Zhengkun Tian ; Zhengqi Wen ; Shuai Zhang
COMMENTS:       submitted to INTERSPEECH2020
HIGHLIGHT:      To address this issue, we propose a non-autoregressive end-to-end speech recognition system called LASO (listen attentively, and spell once).

68, TITLE:      A Weighted Difference of Anisotropic and Isotropic Total Variation for Relaxed Mumford-Shah Color and Multiphase Image Segmentation
http://arxiv.org/abs/2005.04401
AUTHORS:        Kevin Bui ; Fredrick Park ; Yifei Lou ; Jack Xin
HIGHLIGHT:      To deal with the weighted anisotropic-isotropic TV, we apply the difference-of-convex algorithm (DCA), where the subproblems can be minimized by the primal-dual hybrid gradient method (PDHG).

69, TITLE:      Celeganser: Automated Analysis of Nematode Morphology and Age
http://arxiv.org/abs/2005.04884
AUTHORS:        Linfeng Wang ; Shu Kong ; Zachary Pincus ; Charless Fowlkes
COMMENTS:       Computer Vision for Microscopy Image Analysis (CVMI) 2020
HIGHLIGHT:      In this paper we introduce a pipeline for automated analysis of C. elegans imagery for the purpose of studying life-span, health-span and the underlying genetic determinants of aging.

70, TITLE:      Comment on "No-Reference Video Quality Assessment Based on the Temporal Pooling of Deep Features"
http://arxiv.org/abs/2005.04400
AUTHORS:        Franz Götz-Hahn ; Vlad Hosu ; Dietmar Saupe
HIGHLIGHT:      In this letter we report the results from our careful reimplementations.

71, TITLE:      Photo style transfer with consistency losses
http://arxiv.org/abs/2005.04408
AUTHORS:        Xu Yao ; Gilles Puy ; Patrick Pérez
HIGHLIGHT:      We address the problem of style transfer between two photos and propose a new way to preserve photorealism.

72, TITLE:      Gleason Score Prediction using Deep Learning in Tissue Microarray Image
http://arxiv.org/abs/2005.04886
AUTHORS:        Yi-hong Zhang ; Jing Zhang ; Yang Song ; Chaomin Shen ; Guang Yang
HIGHLIGHT:      We used a pre-trained model of prostate segmentation to increase the accuracy of the Gleason grade segmentation.

73, TITLE:      High Resolution Face Age Editing
http://arxiv.org/abs/2005.04410
AUTHORS:        Xu Yao ; Gilles Puy ; Alasdair Newson ; Yann Gousseau ; Pierre Hellier
HIGHLIGHT:      This is the goal of the present work.

74, TITLE:      The Structured Weighted Violations MIRA
http://arxiv.org/abs/2005.04418
AUTHORS:        Dor Ringel ; Rotem Dror ; Roi Reichart
COMMENTS:       7 pages, 1 figure
HIGHLIGHT:      We present the Structured Weighted Violation MIRA (SWVM), a new structured prediction algorithm that is based on an hybridization between MIRA (Crammer and Singer, 2003) and the structured weighted violations perceptron (SWVP) (Dror and Reichart, 2016).

75, TITLE:      Memory-Augmented Relation Network for Few-Shot Learning
http://arxiv.org/abs/2005.04414
AUTHORS:        Jun He ; Xueliang Liu ; Richang Hong
COMMENTS:       To be submitted to ACM Multimedia 2020
HIGHLIGHT:      In this work, we investigate a new metric-learning method, Memory-Augmented Relation Network (MRN), to explicitly exploit these relationships.

76, TITLE:      Building a Manga Dataset "Manga109" with Annotations for Multimedia Applications
http://arxiv.org/abs/2005.04425
AUTHORS:        Kiyoharu Aizawa ; Azuma Fujimoto ; Atsushi Otsubo ; Toru Ogawa ; Tusuke Matsui ; Koki Tsubota ; Hikaru Ikuta
COMMENTS:       10 pages, 8 figures
HIGHLIGHT:      In this article, we describe the details of the dataset and present a few examples of multimedia processing applications (detection, retrieval, and generation) that apply existing deep learning methods and are made possible by the dataset.

77, TITLE:      Understanding Dynamic Scenes using Graph Convolution Networks
http://arxiv.org/abs/2005.04437
AUTHORS:        Sravan Mylavarapu ; Mahtab Sandhu ; Priyesh Vijayan ; K Madhava Krishna ; Balaraman Ravindran ; Anoop Namboodiri
COMMENTS:       Under Review in IROS 2020
HIGHLIGHT:      We present a novel Multi Relational Graph Convolutional Network (MRGCN) to model on-road vehicle behaviours from a sequence of temporally ordered frames as grabbed by a moving monocular camera.

78, TITLE:      Concurrent Separation Logic Meets Template Games
http://arxiv.org/abs/2005.04453
AUTHORS:        Paul-André Melliès ; Léo Stefanesco
HIGHLIGHT:      In this paper, we establish a deep and unexpected connection between two recent lines of work on concurrent separation logic (CSL) and on template game semantics for differential linear logic (DiLL).

79, TITLE:      Transformer-Based Language Models for Similar Text Retrieval and Ranking
http://arxiv.org/abs/2005.04588
AUTHORS:        Javed Qadrud-Din ; Ashraf Bah Rabiou ; Ryan Walker ; Ravi Soni ; Martin Gajek ; Gabriel Pack ; Akhil Rangaraj
COMMENTS:       5 pages, 2 figures
HIGHLIGHT:      In this paper, we introduce novel approaches for effectively applying neural transformer models to similar text retrieval and ranking without an initial bag-of-words-based step.

80, TITLE:      An Integrated Enhancement Solution for 24-hour Colorful Imaging
http://arxiv.org/abs/2005.04580
AUTHORS:        Feifan Lv ; Yinqiang Zheng ; Yicheng Li ; Feng Lu
COMMENTS:       AAAI 2020 (Oral)
HIGHLIGHT:      In this paper, we propose a novel and integrated enhancement solution that produces clear color images, whether at abundant sunlight daytime or extremely low-light nighttime.

81, TITLE:      Maximal Algorithmic Caliber and Algorithmic Causal Network Inference: General Principles of Real-World General Intelligence?
http://arxiv.org/abs/2005.04589
AUTHORS:        Ben Goertzel
HIGHLIGHT:      Maximal Algorithmic Caliber and Algorithmic Causal Network Inference: General Principles of Real-World General Intelligence?

82, TITLE:      Fuzzy Mutation Embedded Hybrids of Gravitational Search and Particle Swarm Optimization Methods for Engineering Design Problems
http://arxiv.org/abs/2005.04599
AUTHORS:        Devroop Kar ; Manosij Ghosh ; Ritam Guha ; Ram Sarkar ; Laura García-Hernández ; Ajith Abraham
COMMENTS:       33 pages, 18 figures, submitted to Engineering Applications of Artificial Intelligence, Elsevier
HIGHLIGHT:      Hence, to solve this issue we have proposed a fuzzy mutation model for two hybrid versions of PSO and GSA - Gravitational Particle Swarm (GPS) and PSOGSA.

83, TITLE:      Insignificant Choice Polynomial Time
http://arxiv.org/abs/2005.04598
AUTHORS:        Klaus-Dieter Schewe
COMMENTS:       56 pages
HIGHLIGHT:      We use this result for our second contribution showing that PTIME differs from NP.

84, TITLE:      Duality in Persistent Homology of Images
http://arxiv.org/abs/2005.04597
AUTHORS:        Adélie Garin ; Teresa Heiss ; Kelly Maggs ; Bea Bleibe ; Vanessa Robins
COMMENTS:       This is an extended abstract for the SoCG Young Researchers Forum 2020
HIGHLIGHT:      Applied to greyscale digital images, we obtain an algorithm to convert barcodes between the two different (dual) topological models of pixel connectivity.

85, TITLE:      A Hybrid Swarm and Gravitation based feature selection algorithm for Handwritten Indic Script Classification problem
http://arxiv.org/abs/2005.04596
AUTHORS:        Ritam Guha ; Manosij Ghosh ; Pawan Kumar Singh ; Ram Sarkar ; Mita Nasipuri
COMMENTS:       37 pages, 22 figures, submitted to Multimedia Tools and Applications, Springer
HIGHLIGHT:      In our paper, we have addressed this issue by introducing a new FS algorithm, called Hybrid Swarm and Gravitation based FS (HSGFS).

86, TITLE:      Embedded Chaotic Whale Survival Algorithm for Filter-Wrapper Feature Selection
http://arxiv.org/abs/2005.04593
AUTHORS:        Ritam Guha ; Manosij Ghosh ; Shyok Mutsuddi ; Ram Sarkar ; Seyedali Mirjalili
COMMENTS:       28 pages, 6 figures, submitted a minor revision to Soft Computing, Springer
HIGHLIGHT:      In this paper, an embedded version of WOA called Embedded Chaotic Whale Survival Algorithm (ECWSA) has been proposed which uses its wrapper process to achieve high classification accuracy and a filter approach to further refine the selected subset with low computation cost.

87, TITLE:      HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment
http://arxiv.org/abs/2005.05005
AUTHORS:        Lingbo Yang ; Chang Liu ; Pan Wang ; Shanshe Wang ; Peiran Ren ; Siwei Ma ; Wen Gao
HIGHLIGHT:      In this paper, we investigate the more challenging and practical "dual-blind" version of the problem by lifting the requirements on both types of prior, termed as "Face Renovation"(FR).

88, TITLE:      Autonomous Tissue Scanning under Free-Form Motion for Intraoperative Tissue Characterisation
http://arxiv.org/abs/2005.05050
AUTHORS:        Jian Zhan ; Joao Cartucho ; Stamatia Giannarou
COMMENTS:       7 pages, 5 figures, ICRA 2020
HIGHLIGHT:      To eliminate these assumptions, we propose a visual servoing framework for autonomous tissue scanning, able to deal with free-form tissue deformation.

89, TITLE:      On the Transferability of Knowledge among Vehicle Routing Problems by using a Cellular Evolutionary Multitasking
http://arxiv.org/abs/2005.05066
AUTHORS:        Eneko Osaba ; Aritz D. Martinez ; Jesus L. Lobo ; Ibai Laña ; Javier Del Ser
COMMENTS:       8 pages, 1 figure, paper accepted for presentation in the 23rd IEEE International Conference on Intelligent Transportation Systems 2020 (IEEE ITSC 2020)
HIGHLIGHT:      The contribution of this research is twofold.

90, TITLE:      A New Computer-Aided Diagnosis System with Modified Genetic Feature Selection for BI-RADS Classification of Breast Masses in Mammograms
http://arxiv.org/abs/2005.05074
AUTHORS:        Said Boumaraf ; Xiabi Liu ; Chokri Ferkous ; Xiaohong Ma
HIGHLIGHT:      This paper proposes a new and effective computer-aided diagnosis (CAD) system to classify mammographic masses into four assessment categories in BI-RADS.

91, TITLE:      Positional Games and QBF: The Corrective Encoding
http://arxiv.org/abs/2005.05098
AUTHORS:        Valentin Mayer-Eichberger ; Abdallah Saffidine
COMMENTS:       Accepted for publication in the 23rd International Conference on Theory and Applications of Satisfiability Testing (SAT2020)
HIGHLIGHT:      We propose a novel encoding of these games into Quantified Boolean Formulas (QBF) such that a game instance admits a winning strategy for first player if and only if the corresponding formula is true.

92, TITLE:      Conditional Image Generation and Manipulation for User-Specified Content
http://arxiv.org/abs/2005.04909
AUTHORS:        David Stap ; Maurits Bleeker ; Sarah Ibrahimi ; Maartje ter Hoeve
COMMENTS:       Accepted to the AI for content creation workshop at CVPR 2020
HIGHLIGHT:      To solve this problem, we propose a single pipeline for text-to-image generation and manipulation. Finally, we introduce the CelebTD-HQ dataset, an extension to CelebA-HQ, consisting of faces and corresponding textual descriptions.

93, TITLE:      An Inductive Transfer Learning Approach using Cycle-consistent Adversarial Domain Adaptation with Application to Brain Tumor Segmentation
http://arxiv.org/abs/2005.04906
AUTHORS:        Yuta Tokuoka ; Shuji Suzuki ; Yohei Sugawara
HIGHLIGHT:      In this work, we provide an inductive transfer learning (ITL) approach to adopt the annotation label of the source domain datasets to tasks of the target domain datasets using Cycle-GAN based unsupervised domain adaptation (UDA).

94, TITLE:      Learning to hash with semantic similarity metrics and empirical KL divergence
http://arxiv.org/abs/2005.04917
AUTHORS:        Heikki Arponen ; Tom E. Bishop
COMMENTS:       17 pages, 5 figures
HIGHLIGHT:      We overcome (i) via a novel loss function encouraging the relative hash code distances of learned features to match those derived from their targets.

95, TITLE:      A Logical Characterization of Constant-Depth Circuits over the Reals
http://arxiv.org/abs/2005.04916
AUTHORS:        Timon Barlag ; Heribert Vollmer
COMMENTS:       22 pages, submitted to MFCS 2020
HIGHLIGHT:      In this paper we give an Immerman's Theorem for real-valued computation.

96, TITLE:      Maximizing Information Gain in Partially Observable Environments via Prediction Reward
http://arxiv.org/abs/2005.04912
AUTHORS:        Yash Satsangi ; Sungsu Lim ; Shimon Whiteson ; Frans Oliehoek ; Martha White
HIGHLIGHT:      Based on this insight we present deep anticipatory networks (DANs), which enables an agent to take actions to reduce its uncertainty without performing explicit belief inference.

97, TITLE:      Towards logical negation for compositional distributional semantics
http://arxiv.org/abs/2005.04929
AUTHORS:        Martha Lewis
HIGHLIGHT:      This paper gives some steps towards providing this operator, modelling it as a version of projection onto the subspace orthogonal to a word.

98, TITLE:      A Deep Learning Approach for Automatic Detection of Fake News
http://arxiv.org/abs/2005.04938
AUTHORS:        Tanik Saikh ; Arkadipta De ; Asif Ekbal ; Pushpak Bhattacharyya
HIGHLIGHT:      In this paper, we propose two effective models based on deep learning for solving fake news detection problem in online news contents of multiple domains.

99, TITLE:      Designing for Human Rights in AI
http://arxiv.org/abs/2005.04949
AUTHORS:        Evgeni Aizenberg ; Jeroen van den Hoven
COMMENTS:       30 pages, 2 figures, pre-print version of the paper currently under review for considering publication
HIGHLIGHT:      In this paper, we bridge this divide through the framework of Design for Values, drawing on methodologies of Value Sensitive Design and Participatory Design to present a roadmap for proactively engaging societal stakeholders to translate fundamental human rights into context-dependent design requirements through a structured, inclusive, and transparent process.

100, TITLE:     Fake Face Detection via Adaptive Residuals Extraction Network
http://arxiv.org/abs/2005.04945
AUTHORS:        Zhiqing Guo ; Gaobo Yang ; Jiyou Chen ; Xingming Sun
HIGHLIGHT:      We propose an adaptive residuals extraction network (AREN), which serves as pre-processing to suppress image content and highlight tampering artifacts.

101, TITLE:     On Rational and Hypergeometric Solutions of Linear Ordinary Difference Equations in $Π\mathbfΣ^*$-field extensions
http://arxiv.org/abs/2005.04944
AUTHORS:        Sergei A. Abramov ; Manuel Bronstein ; Marko Petkovšek ; Carsten Schneider
HIGHLIGHT:      We present a complete algorithm that computes all hypergeometric solutions of homogeneous linear difference equations and rational solutions of parameterized linear difference equations in the setting of $\Pi\Sigma^*$-fields.

102, TITLE:     Propagation Graph Estimation by Pairwise Alignment of Time Series Observation Sequences
http://arxiv.org/abs/2005.04954
AUTHORS:        Tatsuya Hayashi ; Atsuyoshi Nakamura
COMMENTS:       7 pages, 6 figures
HIGHLIGHT:      In this paper, we study the problem of estimating the firing propagation order of cells from the $\{0,1 \}$-state sequences of all the cells, where '1' at the $i$-th position means the firing state of the cell at time step $i$.

103, TITLE:     Quantitative Analysis of Image Classification Techniques for Memory-Constrained Devices
http://arxiv.org/abs/2005.04968
AUTHORS:        Sebastian Müksch ; Theo Olausson ; John Wilhelm ; Pavlos Andreadis
COMMENTS:       8 pages (excluding bibliography), 4 figures, 6 tables
HIGHLIGHT:      This paper presents a comprehensive analysis that shows that even in memory-constrained environments, CNNs implemented memory-optimally using Direct Convolutions outperform ProtoNN, Bonsai and FastGRNN models on 3-channel image classification using CIFAR-10.

104, TITLE:     Prototypical Contrastive Learning of Unsupervised Representations
http://arxiv.org/abs/2005.04966
AUTHORS:        Junnan Li ; Pan Zhou ; Caiming Xiong ; Richard Socher ; Steven C. H. Hoi
HIGHLIGHT:      This paper presents Prototypical Contrastive Learning (PCL), an unsupervised representation learning method that addresses the fundamental limitations of the popular instance-wise contrastive learning.

105, TITLE:     Towards Efficient Normalizers of Primitive Groups
http://arxiv.org/abs/2005.04979
AUTHORS:        Sergio Siccha
HIGHLIGHT:      We present the ideas behind an algorithm to compute normalizers of primitive groups with non-regular socle in polynomial time.

106, TITLE:     Deep Reinforcement Learning for Organ Localization in CT
http://arxiv.org/abs/2005.04974
AUTHORS:        Fernando Navarro ; Anjany Sekuboyina ; Diana Waldmannstetter ; Jan C. Peeken ; Stephanie E. Combs ; Bjoern H. Menze
COMMENTS:       Accepted paper in MIDL 2020
HIGHLIGHT:      In this work, an artificial agent is actively self-taught to localize organs in CT by learning from its asserts and mistakes. Within the context of reinforcement learning, we propose a novel set of actions tailored for organ localization in CT.

107, TITLE:     Prior choice affects ability of Bayesian neural networks to identify unknowns
http://arxiv.org/abs/2005.04987
AUTHORS:        Daniele Silvestro ; Tobias Andermann
HIGHLIGHT:      Here, we explore the effects of different prior distributions on classification tasks in BNNs and evaluate the evidence supporting the predictions based on posterior probabilities approximated by Markov Chain Monte Carlo sampling and by computing Bayes factors.

108, TITLE:     Finding Universal Grammatical Relations in Multilingual BERT
http://arxiv.org/abs/2005.04511
AUTHORS:        Ethan A. Chi ; John Hewitt ; Christopher D. Manning
COMMENTS:       To appear in ACL 2020
HIGHLIGHT:      Motivated by these results, we present an unsupervised analysis method that provides evidence mBERT learns representations of syntactic dependency labels, in the form of clusters which largely agree with the Universal Dependencies taxonomy.

109, TITLE:     What Was Written vs. Who Read It: News Media Profiling Using Text Analysis and Social Media Context
http://arxiv.org/abs/2005.04518
AUTHORS:        Ramy Baly ; Georgi Karadzhov ; Jisun An ; Haewoon Kwak ; Yoan Dinkov ; Ahmed Ali ; James Glass ; Preslav Nakov
COMMENTS:       Factuality of reporting, fact-checking, political ideology, media bias, disinformation, propaganda, social media, news media
HIGHLIGHT:      This approach makes it possible to detect likely "fake news" the moment they are published, by simply checking the reliability of their source.

110, TITLE:     In-memory eigenvector computation in time O(1)
http://arxiv.org/abs/2005.04531
AUTHORS:        Zhong Sun ; Giacomo Pedretti ; Elia Ambrosi ; Alessandro Bricalli ; Daniele Ielmini
COMMENTS:       Accepted by Adv. Intell. Syst
HIGHLIGHT:      In this work, time complexity of the eigenvector computation is investigated, based on the feedback analysis of the crosspoint circuit.

111, TITLE:     Time complexity of in-memory solution of linear systems
http://arxiv.org/abs/2005.04530
AUTHORS:        Zhong Sun ; Giacomo Pedretti ; Piergiulio Mannocci ; Elia Ambrosi ; Alessandro Bricalli ; Daniele Ielmini
COMMENTS:       Accepted by IEEE Trans. Electron Devices. The authors thank Scott Aaronson for helpful discussion about time complexity
HIGHLIGHT:      Based on the theory of feedback circuits, we study the dynamics of the solution of linear systems within a memory array, showing that the time complexity of the solution is free of any direct dependence on the problem size N, rather it is governed by the minimal eigenvalue of an associated matrix of the coefficient matrix.

112, TITLE:     Accelerating Deep Neuroevolution on Distributed FPGAs for Reinforcement Learning Problems
http://arxiv.org/abs/2005.04536
AUTHORS:        Alexis Asseman ; Nicolas Antoine ; Ahmet S. Ozcan
COMMENTS:       12 pages. Submitted to ACM Journal on Emerging Technologies in Computing Systems: Special Issue on Hardware and Algorithms for Efficient Machine Learning
HIGHLIGHT:      Recently, alternative approaches such as evolutionary strategies and deep neuroevolution demonstrated competitive results with faster training time on distributed CPU cores.

113, TITLE:     Article citation study: Context enhanced citation sentiment detection
http://arxiv.org/abs/2005.04534
AUTHORS:        Vishal Vyas ; Kumar Ravi ; Vadlamani Ravi ; V. Uma ; Srirangaraj Setlur ; Venu Govindaraju
COMMENTS:       39 pages, 12 Tables, 5 Figures, Journal Paper
HIGHLIGHT:      For citation analysis, we developed eight datasets comprising citation sentences, which are manually annotated by us into three sentiment polarities viz. positive, negative, and neutral.

114, TITLE:     Unified Models of Human Behavioral Agents in Bandits, Contextual Bandits and RL
http://arxiv.org/abs/2005.04544
AUTHORS:        Baihan Lin ; Guillermo Cecchi ; Djallel Bouneffouf ; Jenna Reinen ; Irina Rish
COMMENTS:       This article supersedes and extends our work arXiv:1706.02897 (MAB) and arXiv:1906.11286 (RL) into the Contextual Bandit (CB) framework. It generalized extensively into multi-armed bandits, contextual bandits and RL settings to create a unified framework of human behavioral agents
HIGHLIGHT:      Motivated by clinical literature of a wide range of neurological and psychiatric disorders, we propose here a more general and flexible parametric framework for sequential decision making that involves a two-stream reward processing mechanism.

115, TITLE:     Replication Markets: Results, Lessons, Challenges and Opportunities in AI Replication
http://arxiv.org/abs/2005.04543
AUTHORS:        Yang Liu ; Michael Gordon ; Juntao Wang ; Michael Bishop ; Yiling Chen ; Thomas Pfeiffer ; Charles Twardy ; Domenico Viganola
COMMENTS:       Appeared at AAAI workshop on Reproducible AI (RAI), 2020
HIGHLIGHT:      These projects were driven by theoretical and conceptual concerns about a high fraction of "false positives" in the scientific publications (Ioannidis, 2005) (and a high prevalence of "questionable research practices" (Simmons, Nelson, and Simonsohn, 2011).

116, TITLE:     A Robust Matching Pursuit Algorithm Using Information Theoretic Learning
http://arxiv.org/abs/2005.04541
AUTHORS:        Miaohua Zhang ; Yongsheng Gao ; Changming Sun ; Michael Blumenstein
COMMENTS:       Accepted by "Pattern Recognition"
HIGHLIGHT:      To overcome these problems, a new OMP algorithm is developed based on the information theoretic learning (ITL), which is built on the following new techniques: (1) an ITL-based correlation (ITL-Correlation) is developed as a new similarity measure which can better exploit higher-order statistics of the data, and is robust against many different types of noise and outliers in a sparse representation framework; (2) a non-second order statistic measurement and minimization method is developed to improve the robustness of OMP by overcoming the limitation of Gaussianity inherent in cost function based on second-order moments.

117, TITLE:     Epipolar Transformers
http://arxiv.org/abs/2005.04551
AUTHORS:        Yihui He ; Rui Yan ; Katerina Fragkiadaki ; Shoou-I Yu
COMMENTS:       CVPR 2020
HIGHLIGHT:      Therefore, we propose the differentiable "epipolar transformer", which enables the 2D detector to leverage 3D-aware features to improve 2D pose estimation.

118, TITLE:     Compact Neural Representation Using Attentive Network Pruning
http://arxiv.org/abs/2005.04559
AUTHORS:        Mahdi Biparva ; John Tsotsos
HIGHLIGHT:      In this work, we describe a Top-Down attention mechanism that is added to a Bottom-Up feedforward network to select important connections and subsequently prune redundant ones at all parametric layers.

119, TITLE:     Class-Aware Domain Adaptation for Improving Adversarial Robustness
http://arxiv.org/abs/2005.04564
AUTHORS:        Xianxu Hou ; Jingxin Liu ; Bolei Xu ; Xiaolong Wang ; Bozhi Liu ; Guoping Qiu
HIGHLIGHT:      To this end, we propose a novel Class-Aware Domain Adaptation (CADA) method for adversarial defense without directly applying adversarial training.

120, TITLE:     Efficient Privacy Preserving Edge Computing Framework for Image Classification
http://arxiv.org/abs/2005.04563
AUTHORS:        Omobayode Fagbohungbe ; Sheikh Rufsan Reza ; Xishuang Dong ; Lijun Qian
HIGHLIGHT:      To address these challenges, a novel privacy preserving edge computing framework is proposed in this paper for image classification.

121, TITLE:     Posterior Control of Blackbox Generation
http://arxiv.org/abs/2005.04560
AUTHORS:        Xiang Lisa Li ; Alexander M. Rush
COMMENTS:       Accepted for publication at ACL 2020
HIGHLIGHT:      In this work, we consider augmenting neural generation models with discrete control states learned through a structured latent-variable approach.

122, TITLE:     Non-recurrent Traffic Congestion Detection with a Coupled Scalable Bayesian Robust Tensor Factorization Model
http://arxiv.org/abs/2005.04567
AUTHORS:        Qin Li ; Huachun Tan ; Xizhu Jiang ; Yuankai Wu ; Linhui Ye
HIGHLIGHT:      In this paper, we present a novel analytical training-free framework based on coupled scalable Bayesian robust tensor factorization (Coupled SBRTF).

123, TITLE:     A Survey on Deep Learning for Neuroimaging-based Brain Disorder Analysis
http://arxiv.org/abs/2005.04573
AUTHORS:        Li Zhang ; Mingliang Wang ; Mingxia Liu ; Daoqiang Zhang
COMMENTS:       30 pages, 7 figures
HIGHLIGHT:      This paper reviews the applications of deep learning methods for neuroimaging-based brain disorder analysis.

124, TITLE:     Belief Rule Based Expert System to Identify the Crime Zones
http://arxiv.org/abs/2005.04570
AUTHORS:        Abhijit Pathak ; Abrar Hossain Tasin
COMMENTS:       6 pages, 7 figures
HIGHLIGHT:      In order to further analyze the crime datasets, the paper introduces an analysis study by combining our findings of the Chittagong crime dataset with demographic information to capture factors that could affect neighborhood safety.

125, TITLE:     Text-Based Ideal Points
http://arxiv.org/abs/2005.04232
AUTHORS:        Keyon Vafa ; Suresh Naidu ; David M. Blei
COMMENTS:       To appear in the Proceedings of the 2020 Conference of the Association for Computational Linguistics (ACL 2020)
HIGHLIGHT:      In this paper, we introduce the text-based ideal point model (TBIP), an unsupervised probabilistic topic model that analyzes texts to quantify the political positions of its authors.

126, TITLE:     ConvoKit: A Toolkit for the Analysis of Conversations
http://arxiv.org/abs/2005.04246
AUTHORS:        Jonathan P. Chang ; Caleb Chiam ; Liye Fu ; Andrew Z. Wang ; Justine Zhang ; Cristian Danescu-Niculescu-Mizil
COMMENTS:       Proceedings of SIGDIAL 2020 (System Demos)
HIGHLIGHT:      This paper describes the design and functionality of ConvoKit, an open-source toolkit for analyzing conversations and the social interactions embedded within.

127, TITLE:     Balancing Objectives in Counseling Conversations: Advancing Forwards or Looking Backwards
http://arxiv.org/abs/2005.04245
AUTHORS:        Justine Zhang ; Cristian Danescu-Niculescu-Mizil
COMMENTS:       14 pages, 6 figures, code available through the Cornell Conversational Analysis Toolkit (https://convokit.cornell.edu/). in Proceedings of ACL, 2020
HIGHLIGHT:      In this work, we develop an unsupervised methodology to quantify how counselors manage this balance.

128, TITLE:     View Invariant Human Body Detection and Pose Estimation from Multiple Depth Sensors
http://arxiv.org/abs/2005.04258
AUTHORS:        Walid Bekhtaoui ; Ruhan Sa ; Brian Teixeira ; Vivek Singh ; Klaus Kirchberg ; Yao-jen Chang ; Ankur Kapoor
HIGHLIGHT:      We propose an end-to-end multi-person 3D pose estimation network, Point R-CNN, using multiple point cloud sources.

129, TITLE:     STINet: Spatio-Temporal-Interactive Network for Pedestrian Detection and Trajectory Prediction
http://arxiv.org/abs/2005.04255
AUTHORS:        Zhishuai Zhang ; Jiyang Gao ; Junhua Mao ; Yukai Liu ; Dragomir Anguelov ; Congcong Li
HIGHLIGHT:      In this work, we present a novel end-to-end two-stage network: Spatio-Temporal-Interactive Network (STINet).

130, TITLE:     VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representation
http://arxiv.org/abs/2005.04259
AUTHORS:        Jiyang Gao ; Chen Sun ; Hang Zhao ; Yi Shen ; Dragomir Anguelov ; Congcong Li ; Cordelia Schmid
COMMENTS:       CVPR 2020
HIGHLIGHT:      To further boost VectorNet's capability in learning context features, we propose a novel auxiliary task to recover the randomly masked out map entities and agent trajectories based on their context.

131, TITLE:     Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics
http://arxiv.org/abs/2005.04269
AUTHORS:        Arsenii Kuznetsov ; Pavel Shvechikov ; Alexander Grishin ; Dmitry Vetrov
COMMENTS:       Under review by the International Conference on Machine Learning
HIGHLIGHT:      This paper investigates a novel way to alleviate the overestimation bias in a continuous control setting.

132, TITLE:     Evaluating Sparse Interpretable Word Embeddings for Biomedical Domain
http://arxiv.org/abs/2005.05114
AUTHORS:        Mohammad Amin Samadi ; Mohammad Sadegh Akhondzadeh ; Sayed Jalal Zahabi ; Mohammad Hossein Manshaei ; Zeinab Maleki ; Payman Adibi
HIGHLIGHT:      We present an inclusive study on interpretability of word embeddings in the medical domain, focusing on the role of sparse methods. For the quantitative evaluation, we introduce an extensive categorized dataset that can be used to quantify interpretability based on category theory. As for the latter, we propose datasets which can be utilized for effective extrinsic evaluation of word vectors in the biomedical domain.

133, TITLE:     FroDO: From Detections to 3D Objects
http://arxiv.org/abs/2005.05125
AUTHORS:        Kejie Li ; Martin Rünz ; Meng Tang ; Lingni Ma ; Chen Kong ; Tanner Schmidt ; Ian Reid ; Lourdes Agapito ; Julian Straub ; Steven Lovegrove ; Richard Newcombe
COMMENTS:       To be published in CVPR 2020. The first two authors contributed equally
HIGHLIGHT:      We introduce FroDO, a method for accurate 3D reconstruction of object instances from RGB video that infers object location, pose and shape in a coarse-to-fine manner.

134, TITLE:     Adversarial Learning for Supervised and Semi-supervised Relation Extraction in Biomedical Literature
http://arxiv.org/abs/2005.04277
AUTHORS:        Peng Su ; K. Vijay-Shanker
HIGHLIGHT:      In this paper, we investigate adversarial training with multiple adversarial examples to benefit the relation extraction task.

135, TITLE:     Fine-Grained Visual Classification with Efficient End-to-end Localization
http://arxiv.org/abs/2005.05123
AUTHORS:        Harald Hanselmann ; Hermann Ney
HIGHLIGHT:      In this work we present an efficient localization module that can be fused with a classification network in an end-to-end setup.

136, TITLE:     New Ideas for Brain Modelling 6
http://arxiv.org/abs/2005.05137
AUTHORS:        Kieran Greer
HIGHLIGHT:      This paper describes implementation details for a 3-level cognitive model, described in the paper series.

137, TITLE:     Incremental Learning for End-to-End Automatic Speech Recognition
http://arxiv.org/abs/2005.04288
AUTHORS:        Li Fu ; Xiaoxiao Li ; Libo Zi
COMMENTS:       5 pages, 2 figures
HIGHLIGHT:      We propose an incremental learning for end-to-end Automatic Speech Recognition (ASR) to extend the model's capacity on a new task while retaining the performance on existing ones.

138, TITLE:     A Contrast-Adaptive Method for Simultaneous Whole-Brain and Lesion Segmentation in Multiple Sclerosis
http://arxiv.org/abs/2005.05135
AUTHORS:        Stefano Cerri ; Oula Puonti ; Dominik S. Meier ; Jens Wuerfel ; Mark Mühlau ; Hartwig R. Siebner ; Koen Van Leemput
HIGHLIGHT:      Here we present a method for the simultaneous segmentation of white matter lesions and normal-appearing neuroanatomical structures from multi-contrast brain MRI scans of multiple sclerosis patients.

139, TITLE:     Extending the Tsetlin Machine With Integer-Weighted Clauses for Increased Interpretability
http://arxiv.org/abs/2005.05131
AUTHORS:        K. Darshana Abeyrathna ; Ole-Christoffer Granmo ; Morten Goodwin
COMMENTS:       20 pages, 10 figures
HIGHLIGHT:      Here, we address the accuracy-interpretability challenge in machine learning by equipping the TM clauses with integer weights.

140, TITLE:     Attentional Bottleneck: Towards an Interpretable Deep Driving Network
http://arxiv.org/abs/2005.04298
AUTHORS:        Jinkyu Kim ; Mayank Bansal
HIGHLIGHT:      We propose an architecture called Attentional Bottleneck with the goal of improving transparency.

141, TITLE:     End-To-End Speech Synthesis Applied to Brazilian Portuguese
http://arxiv.org/abs/2005.05144
AUTHORS:        Edresson Casanova ; Arnaldo Candido Junior ; Frederico Santos de Oliveira ; Christopher Shulby ; João Paulo Teixeira ; Moacir Antonelli Ponti ; Sandra Maria Aluisio
HIGHLIGHT:      In the proposed scenario, a model based on Mozilla TTS and RTISI-LA vocoder presented the best performance, achieving a 4.03 MOS value.

142, TITLE:     An Algorithmic Method of Partial Derivatives
http://arxiv.org/abs/2005.05143
AUTHORS:        Cornelius Brand ; Kevin Pratt
HIGHLIGHT:      We study the following problem and its applications: given a homogeneous degree-$d$ polynomial $g$ as an arithmetic circuit, and a $d \times d$ matrix $X$ whose entries are homogeneous linear polynomials, compute $g(\partial/\partial x_1, \ldots, \partial/\partial x_n) \det X$.

143, TITLE:     Deep Residual Network based food recognition for enhanced Augmented Reality application
http://arxiv.org/abs/2005.04292
AUTHORS:        Siddarth S ; Sainath G
COMMENTS:       Total Pages:7 Total Figures:10
HIGHLIGHT:      Deep neural network based learning approaches is widely utilized for image classification or object detection based problems with remarkable outcomes.

144, TITLE:     Autonomous learning and chaining of motor primitives using the Free Energy Principle
http://arxiv.org/abs/2005.05151
AUTHORS:        Louis Annabi ; Alexandre Pitti ; Mathias Quoy
HIGHLIGHT:      In this article, we apply the Free-Energy Principle to the question of motor primitives learning.

145, TITLE:     Reference Pose Generation for Visual Localization via Learned Features and View Synthesis
http://arxiv.org/abs/2005.05179
AUTHORS:        Zichao Zhang ; Torsten Sattler ; Davide Scaramuzza
COMMENTS:       23 pages, 14 figures
HIGHLIGHT:      In this work, we propose a semi-automated approach to generate reference poses based on feature matching between renderings of a 3D model and real images via learned features.

146, TITLE:     A Self-Training Method for Machine Reading Comprehension with Soft Evidence Extraction
http://arxiv.org/abs/2005.05189
AUTHORS:        Yilin Niu ; Fangkai Jiao ; Mantong Zhou ; Ting Yao ; Jingfang Xu ; Minlie Huang
COMMENTS:       12 pages, accepted by ACL2020
HIGHLIGHT:      To address this problem, we present a Self-Training method (STM), which supervises the evidence extractor with auto-generated evidence labels in an iterative process.

147, TITLE:     Robust Tensor Decomposition for Image Representation Based on Generalized Correntropy
http://arxiv.org/abs/2005.04605
AUTHORS:        Miaohua Zhang ; Yongsheng Gao ; Changming Sun ; Michael Blumenstein
COMMENTS:       13 pages
HIGHLIGHT:      To overcome this problem, in this paper we propose a new robust tensor decomposition method using generalized correntropy criterion (Corr-Tensor).

148, TITLE:     A Unified Weight Learning and Low-Rank Regression Model for Robust Face Recognition
http://arxiv.org/abs/2005.04619
AUTHORS:        Miaohua Zhang ; Yongsheng Gao ; Fanhua Shang
HIGHLIGHT:      In this paper, we address this problem by a unified sparse weight learning and low-rank approximation regression model and applied it to the robust face recognition in the presence of varying types and levels of corruptions, such as random pixel corruptions and block occlusions, or disguise.

149, TITLE:     MOMBAT: Heart Rate Monitoring from Face Video using Pulse Modeling and Bayesian Tracking
http://arxiv.org/abs/2005.04618
AUTHORS:        Puneet Gupta ; Brojeshwar Bhowmick ; Arpan Pal
HIGHLIGHT:      Subsequently, we introduce a Fourier basis based modeling to reconstruct the cardiovascular pulse signal at the locations containing the poor quality, that is, the locations affected by out-of-plane face movements.

150, TITLE:     Variational Clustering: Leveraging Variational Autoencoders for Image Clustering
http://arxiv.org/abs/2005.04613
AUTHORS:        Vignesh Prasad ; Dipanjan Das ; Brojeshwar Bhowmick
HIGHLIGHT:      Since we wish to efficiently discriminate between different clusters in the data, we propose a method based on VAEs where we use a Gaussian Mixture prior to help cluster the images accurately.

151, TITLE:     How Context Affects Language Models' Factual Predictions
http://arxiv.org/abs/2005.04611
AUTHORS:        Fabio Petroni ; Patrick Lewis ; Aleksandra Piktus ; Tim Rocktäschel ; Yuxiang Wu ; Alexander H. Miller ; Sebastian Riedel
COMMENTS:       accepted at AKBC 2020
HIGHLIGHT:      In this paper, we go a step further and integrate information from a retrieval system with a pre-trained language model in a purely unsupervised way.

152, TITLE:     A Comparison of Few-Shot Learning Methods for Underwater Optical and Sonar Image Classification
http://arxiv.org/abs/2005.04621
AUTHORS:        Mateusz Ochal ; Jose Vazquez ; Yvan Petillot ; Sen Wang
COMMENTS:       Accepted to IEEE OCEANS2020 (Singapore)
HIGHLIGHT:      Finding an algorithm capable of learning from only a few samples could reduce the time spent obtaining and labeling datasets, and accelerate the training of deep-learning models.

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

154, TITLE:     A Simple and Scalable Shape Representation for 3D Reconstruction
http://arxiv.org/abs/2005.04623
AUTHORS:        Mateusz Michalkiewicz ; Eugene Belilovsky ; Mahsa Baktashmotlagh ; Anders Eriksson
COMMENTS:       9 pages plus 3 pages of references. 4 figures
HIGHLIGHT:      In this work, we show that this additional complexity is not necessary, and that we can actually obtain high quality 3D reconstruction using a linear decoder, obtained from principal component analysis on the signed distance function (SDF) of the surface.

155, TITLE:     Learning Context-Based Non-local Entropy Modeling for Image Compression
http://arxiv.org/abs/2005.04661
AUTHORS:        Mu Li ; Kai Zhang ; Wangmeng Zuo ; Radu Timofte ; David Zhang
HIGHLIGHT:      To address this issue, we propose a non-local operation for context modeling by employing the global similarity within the context.

156, TITLE:     Domain Adaptation for Image Dehazing
http://arxiv.org/abs/2005.04668
AUTHORS:        Yuanjie Shao ; Lerenhan Li ; Wenqi Ren ; Changxin Gao ; Nong Sang
COMMENTS:       Accepted by IEEE Conference on Computer Vision and Patten Recognition (CVPR), 2020
HIGHLIGHT:      To address this issue, we propose a domain adaptation paradigm, which consists of an image translation module and two image dehazing modules.

157, TITLE:     A Generalized Kernel Risk Sensitive Loss for Robust Two-Dimensional Singular Value Decomposition
http://arxiv.org/abs/2005.04671
AUTHORS:        Miaohua Zhang ; Yongsheng Gao
HIGHLIGHT:      To overcome this problem, we propose a robust 2DSVD framework based on a generalized kernel risk sensitive loss (GKRSL-2DSVD) which is more robust to noise and and outliers.

158, TITLE:     From Standard Summarization to New Tasks and Beyond: Summarization with Manifold Information
http://arxiv.org/abs/2005.04684
AUTHORS:        Shen Gao ; Xiuying Chen ; Zhaochun Ren ; Dongyan Zhao ; Rui Yan
COMMENTS:       Accepted by IJCAI 2020 Survey Track
HIGHLIGHT:      Text summarization is the research area aiming at creating a short and condensed version of the original document, which conveys the main idea of the document in a few words.

159, TITLE:     Segmentation of Macular Edema Datasets with Small Residual 3D U-Net Architectures
http://arxiv.org/abs/2005.04697
AUTHORS:        Jonathan Frawley ; Chris G. Willcocks ; Maged Habib ; Caspar Geenen ; David H. Steel ; Boguslaw Obara
COMMENTS:       7 pages, 5 figures
HIGHLIGHT:      This paper investigates the application of deep convolutional neural networks with prohibitively small datasets to the problem of macular edema segmentation.

160, TITLE:     Non-Autoregressive Image Captioning with Counterfactuals-Critical Multi-Agent Learning
http://arxiv.org/abs/2005.04690
AUTHORS:        Longteng Guo ; Jing Liu ; Xinxin Zhu ; Xingjian He ; Jie Jiang ; Hanqing Lu
COMMENTS:       IJCAI 2020 (copyright held by IJCAI)
HIGHLIGHT:      In this paper, we propose a Non-Autoregressive Image Captioning (NAIC) model with a novel training paradigm: Counterfactuals-critical Multi-Agent Learning (CMAL).



==========Updates to Previous Papers==========
1, TITLE:       A Better Variant of Self-Critical Sequence Training
http://arxiv.org/abs/2003.09971
AUTHORS:        Ruotian Luo
HIGHLIGHT:      In this work, we present a simple yet better variant of Self-Critical Sequence Training.

2, TITLE:       Multi-object Monocular SLAM for Dynamic Environments
http://arxiv.org/abs/2002.03528
AUTHORS:        Gokul B. Nair ; Swapnil Daga ; Rahul Sajnani ; Anirudha Ramesh ; Junaid Ahmed Ansari ; Krishna Murthy Jatavallabhula ; K. Madhava Krishna
COMMENTS:       Accepted to IEEE Intelligent Vehicles Symposium 2020 (IV2020)
HIGHLIGHT:      In this paper, we tackle the problem of multibody SLAM from a monocular camera.

3, TITLE:       Actor Conditioned Attention Maps for Video Action Detection
http://arxiv.org/abs/1812.11631
AUTHORS:        Oytun Ulutan ; Swati Rallapalli ; Mudhakar Srivatsa ; Carlos Torres ; B. S. Manjunath
COMMENTS:       WACV2020 Paper
HIGHLIGHT:      To this end, we propose to replace region of interest(RoI) pooling with an attention module, which ranks each spatio-temporal region's relevance to a detected actor instead of cropping.

4, TITLE:       Prototype Refinement Network for Few-Shot Segmentation
http://arxiv.org/abs/2002.03579
AUTHORS:        Jinlu Liu ; Yongqiang Qin
HIGHLIGHT:      In this paper, we propose a Prototype Refinement Network (PRNet) to attack the challenge of few-shot segmentation.

5, TITLE:       CovidCTNet: An Open-Source Deep Learning Approach to Identify Covid-19 Using CT Image
http://arxiv.org/abs/2005.03059
AUTHORS:        Tahereh Javaheri ; Morteza Homayounfar ; Zohreh Amoozgar ; Reza Reiazi ; Fatemeh Homayounieh ; Engy Abbas ; Azadeh Laali ; Amir Reza Radmard ; Mohammad Hadi Gharib ; Seyed Ali Javad Mousavi ; Omid Ghaemi ; Rosa Babaei ; Hadi Karimi Mobin ; Mehdi Hosseinzadeh ; Rana Jahanban-Esfahlan ; Khaled Seidi ; Mannudeep K. Kalra ; Guanglan Zhang ; L. T. Chitkushev ; Benjamin Haibe-Kains ; Reza Malekzadeh ; Reza Rawassizadeh
COMMENTS:       5 figures
HIGHLIGHT:      To enhance the accuracy of CT imaging detection, we developed an open-source set of algorithms called CovidCTNet that successfully differentiates Covid-19 from community-acquired pneumonia (CAP) and other lung diseases.

6, TITLE:       Meta-Meta-Classification for One-Shot Learning
http://arxiv.org/abs/2004.08083
AUTHORS:        Arkabandhu Chowdhury ; Dipak Chaudhari ; Swarat Chaudhuri ; Chris Jermaine
COMMENTS:       8 pages without references, 2 figures
HIGHLIGHT:      We present a new approach, called meta-meta-classification, to learning in small-data settings.

7, TITLE:       Categorical Vector Space Semantics for Lambek Calculus with a Relevant Modality
http://arxiv.org/abs/2005.03074
AUTHORS:        Lachlan McPheat ; Mehrnoosh Sadrzadeh ; Hadi Wazni ; Gijs Wijnholds
HIGHLIGHT:      We apply the model to construct categorical and concrete semantic interpretations for the motivating example of !

8, TITLE:       Co-Saliency Spatio-Temporal Interaction Network for Person Re-Identification in Videos
http://arxiv.org/abs/2004.04979
AUTHORS:        Jiawei Liu ; Zheng-Jun Zha ; Xierong Zhu ; Na Jiang
HIGHLIGHT:      In this work, we propose a novel Co-Saliency Spatio-Temporal Interaction Network (CSTNet) for person re-identification in videos.

9, TITLE:       Few-Shot Object Detection with Attention-RPN and Multi-Relation Detector
http://arxiv.org/abs/1908.01998
AUTHORS:        Qi Fan ; Wei Zhuo ; Chi-Keung Tang ; Yu-Wing Tai
COMMENTS:       CVPR2020 Camera Ready. (Fix Figure 3 and Table 5. More implementation details in the supplementary material.)
HIGHLIGHT:      In this paper, we propose a novel few-shot object detection network that aims at detecting objects of unseen categories with only a few annotated examples.

10, TITLE:      Inference with Choice Functions Made Practical
http://arxiv.org/abs/2005.03098
AUTHORS:        Arne Decadt ; Jasper De Bock ; Gert de Cooman
HIGHLIGHT:      We present a practical algorithm to compute this natural extension and provide several methods that can be used to improve its scalability.

11, TITLE:      Derivation of a Constant Velocity Motion Model for Visual Tracking
http://arxiv.org/abs/2005.00844
AUTHORS:        Nathanael L. Baisa
HIGHLIGHT:      In this document, we derive the constant velocity motion model that incorporates sizes of objects that, we think, can help the new researchers to adapt to it very quickly.

12, TITLE:      A Novel Deep Learning Pipeline for Retinal Vessel Detection in Fluorescein Angiography
http://arxiv.org/abs/1907.02946
AUTHORS:        Li Ding ; Mohammad H. Bawany ; Ajay E. Kuriyan ; Rajeev S. Ramchandran ; Charles C. Wykoff ; Gaurav Sharma
COMMENTS:       A paper based on this pre-print has been published (after revisions) in IEEE Trans. Image Processing. See the first footnote on front page of the article for details
HIGHLIGHT:      We propose a novel pipeline to detect retinal vessels in FA images using deep neural networks that reduces the effort required for generating labeled ground truth data by combining two key components: cross-modality transfer and human-in-the-loop learning.

13, TITLE:      Synetgy: Algorithm-hardware Co-design for ConvNet Accelerators on Embedded FPGAs
http://arxiv.org/abs/1811.08634
AUTHORS:        Yifan Yang ; Qijing Huang ; Bichen Wu ; Tianjun Zhang ; Liang Ma ; Giulio Gambardella ; Michaela Blott ; Luciano Lavagno ; Kees Vissers ; John Wawrzynek ; Kurt Keutzer
COMMENTS:       Update to the latest results
HIGHLIGHT:      In this work, we adopt an algorithm-hardware co-design approach to develop a ConvNet accelerator called Synetgy and a novel ConvNet model called DiracDeltaNet$^{\dagger}$.

14, TITLE:      Interpreting Verbal Irony: Linguistic Strategies and the Connection to the Type of Semantic Incongruity
http://arxiv.org/abs/1911.00891
AUTHORS:        Debanjan Ghosh ; Elena Musi ; Kartikeya Upasani ; Smaranda Muresan
COMMENTS:       Accepted at Society for Computation in Linguistics (SCiL), 2020 Conference
HIGHLIGHT:      We propose a typology of linguistic strategies for verbal irony interpretation and link it to various theoretical linguistic frameworks.

15, TITLE:      BreizhCrops: A Time Series Dataset for Crop Type Mapping
http://arxiv.org/abs/1905.11893
AUTHORS:        Marc Rußwurm ; Charlotte Pelletier ; Maximilian Zollner ; Sébastien Lefèvre ; Marco Körner
COMMENTS:       accepted to ISPRS Archives 2020
HIGHLIGHT:      We present Breizhcrops, a novel benchmark dataset for the supervised classification of field crops from satellite time series.

16, TITLE:      DeepFakes and Beyond: A Survey of Face Manipulation and Fake Detection
http://arxiv.org/abs/2001.00179
AUTHORS:        Ruben Tolosana ; Ruben Vera-Rodriguez ; Julian Fierrez ; Aythami Morales ; Javier Ortega-Garcia
HIGHLIGHT:      Among all the aspects discussed in the survey, we pay special attention to the latest generation of DeepFakes, highlighting its improvements and challenges for fake detection.

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

18, TITLE:      KinGDOM: Knowledge-Guided DOMain adaptation for sentiment analysis
http://arxiv.org/abs/2005.00791
AUTHORS:        Deepanway Ghosal ; Devamanyu Hazarika ; Abhinaba Roy ; Navonil Majumder ; Rada Mihalcea ; Soujanya Poria
HIGHLIGHT:      In this paper, we take a novel perspective on this task by exploring the role of external commonsense knowledge.

19, TITLE:      Robust Motion Averaging under Maximum Correntropy Criterion
http://arxiv.org/abs/2004.09829
AUTHORS:        Jihua Zhu ; Jie Hu ; Huimin Lu ; Badong Chen ; Zhongyu Li
HIGHLIGHT:      Accordingly, this paper proposes a novel robust motion averaging method based on the maximum correntropy criterion (MCC).

20, TITLE:      Autoencoding Pixies: Amortised Variational Inference with Graph Convolutions for Functional Distributional Semantics
http://arxiv.org/abs/2005.02991
AUTHORS:        Guy Emerson
COMMENTS:       To be published in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL); added acknowledgements
HIGHLIGHT:      In this paper, I introduce the Pixie Autoencoder, which augments the generative model of Functional Distributional Semantics with a graph-convolutional neural network to perform amortised variational inference.

21, TITLE:      Synchronous Bidirectional Learning for Multilingual Lip Reading
http://arxiv.org/abs/2005.03846
AUTHORS:        Mingshuang Luo ; Shuang Yang ; Xilin Chen ; Zitao Liu ; Shiguang Shan
COMMENTS:       12 pages,2 figures,4 tables
HIGHLIGHT:      To make the learning process more targeted at each particular language, we introduce an extra task of predicting the language identity in the learning process.

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:      Combining Visual and Textual Features for Semantic Segmentation of Historical Newspapers
http://arxiv.org/abs/2002.06144
AUTHORS:        Raphaël Barman ; Maud Ehrmann ; Simon Clematide ; Sofia Ares Oliveira ; Frédéric Kaplan
HIGHLIGHT:      In this context, we introduce a multimodal approach for the semantic segmentation of historical newspapers that combines visual and textual features.

24, TITLE:      MCMC assisted by Belief Propagation
http://arxiv.org/abs/1605.09042
AUTHORS:        Sungsoo Ahn ; Michael Chertkov ; Jinwoo Shin
COMMENTS:       Fixed minor typos
HIGHLIGHT:      In this paper, we introduce MCMC algorithms correcting the approximation error of BP, i.e., we provide a way to compensate for BP errors via a consecutive BP-aware MCMC.

25, TITLE:      NTIRE 2020 Challenge on Image and Video Deblurring
http://arxiv.org/abs/2005.01244
AUTHORS:        Seungjun Nah ; Sanghyun Son ; Radu Timofte ; Kyoung Mu Lee
COMMENTS:       To be published in CVPR 2020 Workshop (New Trends in Image Restoration and Enhancement)
HIGHLIGHT:      In this challenge, we present the evaluation results from 3 competition tracks as well as the proposed solutions.

26, TITLE:      A combination of 'pooling' with a prediction model can reduce by 73% the number of COVID-19 (Corona-virus) tests
http://arxiv.org/abs/2005.03453
AUTHORS:        Tomer Cohen ; Lior Finkelman ; Gal Grimberg ; Gadi Shenhar ; Ofer Strichman ; Yonatan Strichman ; Stav Yeger
HIGHLIGHT:      We show that combining a prediction model (based on neural networks), with a new method of test pooling (better than the original Dorfman method, and better than double-pooling) called 'Grid', we can reduce the number of Covid-19 tests by 73%.

27, TITLE:      GIMP-ML: Python Plugins for using Computer Vision Models in GIMP
http://arxiv.org/abs/2004.13060
AUTHORS:        Kritik Soman
COMMENTS:       5 pages, 2 figures
HIGHLIGHT:      This paper introduces GIMP-ML, a set of Python plugins for the widely popular GNU Image Manipulation Program (GIMP).

28, TITLE:      Ethics in the digital era
http://arxiv.org/abs/2003.06530
AUTHORS:        David Pastor-Escuredo
HIGHLIGHT:      Ethics in the digital era

29, TITLE:      Influence of segmentation on deep iris recognition performance
http://arxiv.org/abs/1901.10431
AUTHORS:        Juš Lozej ; Dejan Štepec ; Vitomir Štruc ; Peter Peer
COMMENTS:       6 pages, 3 figures, 3 tables, submitted to IWBF 2019
HIGHLIGHT:      To answer this question, we present in this paper an empirical analysis of the impact of iris segmentation on the performance of deep learning models using a simple two stage pipeline consisting of a segmentation and a recognition step.

30, TITLE:      Attention-based Assisted Excitation for Salient Object Detection
http://arxiv.org/abs/2003.14194
AUTHORS:        Saeed Masoudnia ; Melika Kheirieh ; Abdol-Hossein Vahabie ; Babak Nadjar Araabi
HIGHLIGHT:      In this paper, object-based attention in human visual cortex inspires us to introduce a mechanism for modification of activations in feature maps of CNNs.

31, TITLE:      ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context
http://arxiv.org/abs/2005.03191
AUTHORS:        Wei Han ; Zhengdong Zhang ; Yu Zhang ; Jiahui Yu ; Chung-Cheng Chiu ; James Qin ; Anmol Gulati ; Ruoming Pang ; Yonghui Wu
HIGHLIGHT:      In this paper, we study how to bridge this gap and go beyond with a novel CNN-RNN-transducer architecture, which we call ContextNet.

32, TITLE:      Understanding the Stability of Medical Concept Embeddings
http://arxiv.org/abs/1904.09552
AUTHORS:        Grace E. Lee ; Aixin Sun
COMMENTS:       v2
HIGHLIGHT:      In this work, we conduct a detailed analysis on the stability of concept embeddings in medical domain, particularly the relation with concept frequency.

33, TITLE:      A Computationally Efficient Pipeline Approach to Full Page Offline Handwritten Text Recognition
http://arxiv.org/abs/1910.00663
AUTHORS:        Jonathan Chung ; Thomas Delteil
HIGHLIGHT:      In this paper, a less computationally expensive full page offline handwritten text recognition framework is introduced.

34, TITLE:      Multi-Task Learning Enhanced Single Image De-Raining
http://arxiv.org/abs/2003.09689
AUTHORS:        Yulong Fan ; Rong Chen ; Bo Li
COMMENTS:       15 pages, 16 figures
HIGHLIGHT:      In this paper, we address a non-trivial issue of removing visual effect of rain streak from a single image.

35, TITLE:      Analysis of Deep Complex-Valued Convolutional Neural Networks for MRI Reconstruction
http://arxiv.org/abs/2004.01738
AUTHORS:        Elizabeth K. Cole ; Joseph Y. Cheng ; John M. Pauly ; Shreyas S. Vasanawala
HIGHLIGHT:      In this work, we investigate end-to-end complex-valued convolutional neural networks - specifically, for image reconstruction in lieu of two-channel real-valued networks.

36, TITLE:      Prediction of Dashed Café Wall illusion by the Classical Receptive Field Model
http://arxiv.org/abs/1902.03739
AUTHORS:        Nasim Nematzadeh ; David M. W. Powers
COMMENTS:       6 pages, 6 figures. Accepted in ICECCE2020 conference
HIGHLIGHT:      We demonstrate that a simple Differences of Gaussian model provides an explanatory mechanism for the illusory tilt perceived in a family of Caf\'e Wall illusion generalizes to the dashed versions of Caf\'e Wall.

37, TITLE:      Maximum Absolute Determinants of Upper Hessenberg Bohemian Matrices
http://arxiv.org/abs/2003.00454
AUTHORS:        Jonathan P. Keating ; Ahmet Abdullah Keleş
HIGHLIGHT:      We determine the maximum absolute determinant of upper Hessenberg Bohemian Matrices for which the subdiagonal entries are fixed to be $1$ and upper triangular entries are sampled from $\{0,1,\cdots,n\}$, extending previous results for $n=1$ and $n=2$ and proving a recent conjecture of Fasi & Negri Porzio [8].

38, TITLE:      Speeding up Word Mover's Distance and its variants via properties of distances between embeddings
http://arxiv.org/abs/1912.00509
AUTHORS:        Matheus Werner ; Eduardo Laber
COMMENTS:       Author's final version, accepted for publication at ECAI2020
HIGHLIGHT:      Relying on assumptions that are supported by empirical properties of the distances between embeddings, we propose an approach to speed up both WMD and RWMD.

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

40, TITLE:      Noisy-As-Clean: Learning Self-supervised Denoising from the Corrupted Image
http://arxiv.org/abs/1906.06878
AUTHORS:        Jun Xu ; Yuan Huang ; Ming-Ming Cheng ; Li Liu ; Fan Zhu ; Zhou Xu ; Ling Shao
COMMENTS:       12 pages, 9 figures, 6 tables, the first two authors contribute equally
HIGHLIGHT:      To circumvent this problem, in this work, we propose a novel "Noisy-As-Clean" (NAC) strategy of training self-supervised denoising networks.

41, TITLE:      Copy mechanism and tailored training for character-based data-to-text generation
http://arxiv.org/abs/1904.11838
AUTHORS:        Marco Roberti ; Giovanni Bonetta ; Rossella Cancelliere ; Patrick Gallinari
COMMENTS:       ECML-PKDD 2019 (Camera ready version)
HIGHLIGHT:      In this work, we present an end-to-end sequence-to-sequence model with attention mechanism which reads and generates at a character level, no longer requiring delexicalization, tokenization, nor even lowercasing. We also introduce a new dataset called E2E+, designed to highlight the copying capabilities of character-based models, that is a modified version of the well-known E2E dataset used in the E2E Challenge.

42, 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.

43, TITLE:      Natural Image Manipulation for Autoregressive Models Using Fisher Scores
http://arxiv.org/abs/1912.05015
AUTHORS:        Wilson Yan ; Jonathan Ho ; Pieter Abbeel
HIGHLIGHT:      In this paper, we propose using Fisher scores as a method to extract embeddings from an autoregressive model to use for interpolation and show that our method provides more meaningful sample manipulation compared to alternate embeddings such as network activations.

44, TITLE:      Fine-Grained Expression Manipulation via Structured Latent Space
http://arxiv.org/abs/2004.09769
AUTHORS:        Junshu Tang ; Zhiwen Shao ; Lizhuang Ma
HIGHLIGHT:      To tackle this limitation, we propose an end-to-end expression-guided generative adversarial network (EGGAN), which utilizes structured latent codes and continuous expression labels as input to generate images with expected expressions.

45, TITLE:      Local Context Normalization: Revisiting Local Normalization
http://arxiv.org/abs/1912.05845
AUTHORS:        Anthony Ortiz ; Caleb Robinson ; Dan Morris ; Olac Fuentes ; Christopher Kiekintveld ; Md Mahmudulla Hassan ; Nebojsa Jojic
COMMENTS:       Accepted as a CVPR 2020 oral paper. arXiv admin note: text overlap with arXiv:1803.08494 by other authors
HIGHLIGHT:      In this paper, we propose Local Context Normalization (LCN): a normalization layer where every feature is normalized based on a window around it and the filters in its group.

46, TITLE:      Reversible Adversarial Example based on Reversible Image Transformation
http://arxiv.org/abs/1911.02360
AUTHORS:        Zhaoxia Yin ; Hua Wang ; Jie Wang ; Weiming Zhang
HIGHLIGHT:      In this paper, we take advantage of Reversible Image Transformation (RIT) to disguise original image as its adversarial example to get a controllable adversarial example, namely reversible adversarial example, which is still an adversarial example to DNNs.

47, TITLE:      The Dynamic Travelling Thief Problem: Benchmarks and Performance of Evolutionary Algorithms
http://arxiv.org/abs/2004.12045
AUTHORS:        Ragav Sachdeva ; Frank Neumann ; Markus Wagner
HIGHLIGHT:      With this article, we define a number of scenarios based on the Travelling Thief Problem to enable research on the effect of dynamic changes to sub-components.

48, TITLE:      UiO-UvA at SemEval-2020 Task 1: Contextualised Embeddings for Lexical Semantic Change Detection
http://arxiv.org/abs/2005.00050
AUTHORS:        Andrey Kutuzov ; Mario Giulianelli
HIGHLIGHT:      We apply contextualised word embeddings to lexical semantic change detection in the SemEval-2020 Shared Task 1.

49, TITLE:      SAIA: Split Artificial Intelligence Architecture for Mobile Healthcare System
http://arxiv.org/abs/2004.12059
AUTHORS:        Di Zhuang ; Nam Nguyen ; Keyu Chen ; J. Morris Chang
COMMENTS:       17 pages, 9 figures, 2 tables
HIGHLIGHT:      We propose SAIA, a Split Artificial Intelligence Architecture for mobile healthcare systems.

50, TITLE:      Learning Implicit Text Generation via Feature Matching
http://arxiv.org/abs/2005.03588
AUTHORS:        Inkit Padhi ; Pierre Dognin ; Ke Bai ; Cicero Nogueira dos Santos ; Vijil Chenthamarakshan ; Youssef Mroueh ; Payel Das
COMMENTS:       ACL 2020
HIGHLIGHT:      In this paper, we present new GFMN formulations that are effective for sequential data.

51, TITLE:      A Safety Framework for Critical Systems Utilising Deep Neural Networks
http://arxiv.org/abs/2003.05311
AUTHORS:        Xingyu Zhao ; Alec Banks ; James Sharp ; Valentin Robu ; David Flynn ; Michael Fisher ; Xiaowei Huang
COMMENTS:       Accepted by SafeComp2020
HIGHLIGHT:      Working towards addressing this challenge, this paper presents a principled novel safety argument framework for critical systems that utilise deep neural networks.

52, TITLE:      A Survey of the Recent Architectures of Deep Convolutional Neural Networks
http://arxiv.org/abs/1901.06032
AUTHORS:        Asifullah Khan ; Anabia Sohail ; Umme Zahoora ; Aqsa Saeed Qureshi
COMMENTS:       Number of Pages: 70, Number of Figures: 11, Number of Tables: 11. Artif Intell Rev (2020)
HIGHLIGHT:      This survey thus focuses on the intrinsic taxonomy present in the recently reported deep CNN architectures and, consequently, classifies the recent innovations in CNN architectures into seven different categories.

53, TITLE:      Intermediate-Task Transfer Learning with Pretrained Models for Natural Language Understanding: When and Why Does It Work?
http://arxiv.org/abs/2005.00628
AUTHORS:        Yada Pruksachatkun ; Jason Phang ; Haokun Liu ; Phu Mon Htut ; Xiaoyi Zhang ; Richard Yuanzhe Pang ; Clara Vania ; Katharina Kann ; Samuel R. Bowman
COMMENTS:       ACL 2020
HIGHLIGHT:      To investigate this, we perform a large-scale study on the pretrained RoBERTa model with 110 intermediate-target task combinations.

54, TITLE:      Reconciling progress-insensitive noninterference and declassification
http://arxiv.org/abs/2005.01977
AUTHORS:        Johan Bay ; Aslan Askarov
HIGHLIGHT:      We believe that the connection established in this work will enable other applications of ideas from the literature on declassification to progress insensitivity.

55, TITLE:      Spectral Characterization of functional MRI data on voxel-resolution cortical graphs
http://arxiv.org/abs/1910.09507
AUTHORS:        Hamid Behjat ; Martin Larsson
COMMENTS:       Fixed two typos in the equations; (1) definition of L in section 2.1, paragraph 1. (2) signal de-meaning and normalization in section 2.4, paragraph 1
HIGHLIGHT:      We present a method to study fMRI data on subject-specific cerebral hemisphere cortex (CHC) graphs, which encode the cortical morphology at the resolution of voxels in 3-D.

56, TITLE:      The Marriage of Univalence and Parametricity
http://arxiv.org/abs/1909.05027
AUTHORS:        Nicolas Tabareau ; Éric Tanter ; Matthieu Sozeau
COMMENTS:       journal revision
HIGHLIGHT:      The Marriage of Univalence and Parametricity

57, TITLE:      Transformation Consistent Self-ensembling Model for Semi-supervised Medical Image Segmentation
http://arxiv.org/abs/1903.00348
AUTHORS:        Xiaomeng Li ; Lequan Yu ; Hao Chen ; Chi-Wing Fu ; Lei Xing ; Pheng-Ann Heng
COMMENTS:       Accept at IEEE Transactions on Neural Networks and Learning Systems
HIGHLIGHT:      In this paper, we present a novel semi-supervised method for medical image segmentation, where the network is optimized by the weighted combination of a common supervised loss for labeled inputs only and a regularization loss for both labeled and unlabeled data.

58, TITLE:      Learning Improvement Heuristics for Solving Routing Problems
http://arxiv.org/abs/1912.05784
AUTHORS:        Yaoxin Wu ; Wen Song ; Zhiguang Cao ; Jie Zhang ; Andrew Lim
COMMENTS:       10 pages, 4 figures
HIGHLIGHT:      In this paper, we propose a deep reinforcement learning framework to learn the improvement heuristics for routing problems.

59, TITLE:      Open-plan Glare Evaluator (OGE): A Demonstration of a New Glare Prediction Approach Using Machine Learning Algorithms
http://arxiv.org/abs/1910.05594
AUTHORS:        Ayman Wagdy ; Veronica Garcia-Hansen ; Mohammed Elhenawy ; Gillian Isoardi ; Robin Drogemuller ; Fatma Fathy
HIGHLIGHT:      This research aims to demonstrate the validity of this approach by comparing the accuracy of the new ML model for open-plan offices (OGE) to the accuracy of the existing glare metrics using local dataset.

60, TITLE:      A Domain Semantics for Higher-Order Recursive Processes
http://arxiv.org/abs/2002.01960
AUTHORS:        Ryan Kavanagh
HIGHLIGHT:      We give polarized SILL a domain-theoretic semantics---the first denotational semantics for a language with this combination of features.

61, TITLE:      Multi-Sentence Argument Linking
http://arxiv.org/abs/1911.03766
AUTHORS:        Seth Ebner ; Patrick Xia ; Ryan Culkin ; Kyle Rawlins ; Benjamin Van Durme
COMMENTS:       ACL 2020
HIGHLIGHT:      We present a novel document-level model for finding argument spans that fill an event's roles, connecting related ideas in sentence-level semantic role labeling and coreference resolution.

62, TITLE:      Towards a Complexity-theoretic Understanding of Restarts in SAT solvers
http://arxiv.org/abs/2003.02323
AUTHORS:        Chunxiao Li ; Noah Fleming ; Marc Vinyals ; Toniann Pitassi ; Vijay Ganesh
HIGHLIGHT:      In this paper, we prove a series of theoretical results that characterize the power of restarts for various models of SAT solvers.

63, TITLE:      Token Manipulation Generative Adversarial Network for Text Generation
http://arxiv.org/abs/2005.02794
AUTHORS:        DaeJin Jo
COMMENTS:       5 pages, 2 figures
HIGHLIGHT:      In this paper, we focus on addressing the limitations caused by having to specify blanks to be filled.

64, TITLE:      The Right Tool for the Job: Matching Model and Instance Complexities
http://arxiv.org/abs/2004.07453
AUTHORS:        Roy Schwartz ; Gabriel Stanovsky ; Swabha Swayamdipta ; Jesse Dodge ; Noah A. Smith
COMMENTS:       ACL 2020; 12 pages; code available in https://github.com/allenai/sledgehammer
HIGHLIGHT:      To better respect a given inference budget, we propose a modification to contextual representation fine-tuning which, during inference, allows for an early (and fast) "exit" from neural network calculations for simple instances, and late (and accurate) exit for hard instances.

65, TITLE:      Enriching Documents with Compact, Representative, Relevant Knowledge Graphs
http://arxiv.org/abs/2005.00153
AUTHORS:        Shuxin Li ; Zixian Huang ; Gong Cheng ; Evgeny Kharlamov ; Kalpa Gunaratna
COMMENTS:       7 pages, accepted to IJCAI-PRICAI 2020. The paper is temporarily withdrawn due to company policies
HIGHLIGHT:      Existing methods identify mentions of entities in a background KG and enrich documents with entity types and direct relations.

66, TITLE:      Neural Entity Summarization with Joint Encoding and Weak Supervision
http://arxiv.org/abs/2005.00152
AUTHORS:        Junyou Li ; Gong Cheng ; Qingxia Liu ; Wen Zhang ; Evgeny Kharlamov ; Kalpa Gunaratna ; Huajun Chen
COMMENTS:       7 pages, accepted to IJCAI-PRICAI 2020 The paper is temporarily withdrawn due to company policies
HIGHLIGHT:      In this paper, we present a supervised approach NEST that is based on our novel neural model to jointly encode graph structure and text in KGs and generate high-quality diversified summaries.

67, TITLE:      Under the Hood of Neural Networks: Characterizing Learned Representations by Functional Neuron Populations and Network Ablations
http://arxiv.org/abs/2004.01254
AUTHORS:        Richard Meyes ; Constantin Waubert de Puiseau ; Andres Posada-Moreno ; Tobias Meisen
COMMENTS:       17 pages, 14 figures, adjusted "under review" statement
HIGHLIGHT:      We address today's lack of transparency of neural networks and shed light on the roles of single neurons and groups of neurons within the network fulfilling a learned task.

68, TITLE:      Soft Threshold Weight Reparameterization for Learnable Sparsity
http://arxiv.org/abs/2002.03231
AUTHORS:        Aditya Kusupati ; Vivek Ramanujan ; Raghav Somani ; Mitchell Wortsman ; Prateek Jain ; Sham Kakade ; Ali Farhadi
COMMENTS:       18 pages, 10 figures
HIGHLIGHT:      This work proposes Soft Threshold Reparameterization (STR), a novel use of the soft-threshold operator on DNN weights.

69, TITLE:      CHAOS Challenge -- Combined (CT-MR) Healthy Abdominal Organ Segmentation
http://arxiv.org/abs/2001.06535
AUTHORS:        A. Emre Kavur ; N. Sinem Gezer ; Mustafa Barış ; Sinem Aslan ; Pierre-Henri Conze ; Vladimir Groza ; Duc Duy Pham ; Soumick Chatterjee ; Philipp Ernst ; Savaş Özkan ; Bora Baydar ; Dmitry Lachinov ; Shuo Han ; Josef Pauli ; Fabian Isensee ; Matthias Perkonigg ; Rachana Sathish ; Ronnie Rajan ; Debdoot Sheet ; Gurbandurdy Dovletov ; Oliver Speck ; Andreas Nürnberger ; Klaus H. Maier-Hein ; Gözde Bozdağı Akar ; Gözde Ünal ; Oğuz Dicle ; M. Alper Selver
COMMENTS:       19 pages, 8 tables, 7 figures
HIGHLIGHT:      The analysis shows that the performance of DL models for single modality (CT / MR) can show reliable volumetric analysis performance (DICE: 0.98 $\pm$ 0.00 / 0.95 $\pm$ 0.01) but the best MSSD performance remain limited (21.89 $\pm$ 13.94 / 20.85 $\pm$ 10.63 mm).

70, TITLE:      Parameterised Counting in Logspace
http://arxiv.org/abs/1904.12156
AUTHORS:        Anselm Haak ; Arne Meier ; Om Prakash ; Raghavendra Rao B. V
COMMENTS:       Changed title to give a better idea of the content at first glance, also many minor corrections and polishing
HIGHLIGHT:      In this article, we study the counting versions of such operators and introduce variants based on tail-nondeterminism, paraW[1] and paraBetaTail, in the setting of parameterised logarithmic space.

71, TITLE:      Multi-domain CT Metal Artifacts Reduction Using Partial Convolution Based Inpainting
http://arxiv.org/abs/1911.05530
AUTHORS:        Artem Pimkin ; Alexander Samoylenko ; Natalia Antipina ; Anna Ovechkina ; Andrey Golanov ; Alexandra Dalechina ; Mikhail Belyaev
HIGHLIGHT:      In this paper, we are exploring the capabilities of a multi-domain method which consists of both sinogram correction (projection domain step) and restored image correction (image-domain step).

72, TITLE:      Explaining Data-Driven Decisions made by AI Systems: The Counterfactual Approach
http://arxiv.org/abs/2001.07417
AUTHORS:        Carlos Fernández-Loría ; Foster Provost ; Xintian Han
HIGHLIGHT:      This paper presents two fundamental reasons why explaining model predictions is not the same as explaining the decisions made using those predictions, suggesting we should carefully consider whether importance-weight explanations are well-suited to explain decisions made by AI systems.

73, TITLE:      Addressing Artificial Intelligence Bias in Retinal Disease Diagnostics
http://arxiv.org/abs/2004.13515
AUTHORS:        Philippe Burlina ; Neil Joshi ; William Paul ; Katia D. Pacheco ; Neil M. Bressler
HIGHLIGHT:      This study evaluated novel AI and deep learning generative methods to address AI bias for retinal diagnostic applications when specifically applied to diabetic retinopathy (DR).

74, TITLE:      Attend to the beginning: A study on using bidirectional attention for extractive summarization
http://arxiv.org/abs/2002.03405
AUTHORS:        Ahmed Magooda ; Cezary Marcjan
COMMENTS:       To be published in FLAIRS33 (https://www.flairs-33.info/) and appear in he proceedings of AAAI
HIGHLIGHT:      In this work, we propose attending to the beginning of a document, to improve the performance of extractive summarization models when applied to forum discussion data.

75, TITLE:      Simple yet Effective Way for Improving the Performance of GAN
http://arxiv.org/abs/1911.10979
AUTHORS:        Yong-Goo Shin ; Yoon-Jae Yeo ; Sung-Jea Ko
COMMENTS:       Submitted to IEEE transactions on neural networks and learning systems
HIGHLIGHT:      To alleviate this problem, this brief presents a simple but effective way that improves the performance of generative adversarial network (GAN) without imposing the training overhead or modifying the network architectures of existing methods.

76, 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.

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

78, TITLE:      Contrast-weighted Dictionary Learning Based Saliency Detection for Remote Sensing Images
http://arxiv.org/abs/2004.02428
AUTHORS:        Zhou Huang ; Huai-Xin Chen ; Tao Zhou ; Yun-Zhi Yang ; Chang-Yin Wang ; Bi-Yuan Liu
HIGHLIGHT:      In this paper, a novel saliency detection model based on Contrast-weighted Dictionary Learning (CDL) is proposed for remote sensing images.

79, TITLE:      Dense RepPoints: Representing Visual Objects with Dense Point Sets
http://arxiv.org/abs/1912.11473
AUTHORS:        Ze Yang ; Yinghao Xu ; Han Xue ; Zheng Zhang ; Raquel Urtasun ; Liwei Wang ; Stephen Lin ; Han Hu
HIGHLIGHT:      We present a new object representation, called Dense RepPoints, which utilize a large number of points to describe the multi-grained object representation of both box level and pixel level.

80, TITLE:      COVID-Net: A Tailored Deep Convolutional Neural Network Design for Detection of COVID-19 Cases from Chest X-Ray Images
http://arxiv.org/abs/2003.09871
AUTHORS:        Linda Wang ; Alexander Wong
COMMENTS:       12 pages
HIGHLIGHT:      Motivated by this and inspired by the open source efforts of the research community, in this study we introduce COVID-Net, a deep convolutional neural network design tailored for the detection of COVID-19 cases from chest X-ray (CXR) images that is open source and available to the general public.

81, TITLE:      Stationary points at infinity for analytic combinatorics
http://arxiv.org/abs/1905.05250
AUTHORS:        Yuliy Baryshnikov ; Stephen Melczer ; Robin Pemantle
COMMENTS:       Updated presentation and proof of main result. Changed notation to differentiate terminology from analytic combinatorics and algebraic geometry
HIGHLIGHT:      Here we establish checkable conditions under which the behavior at infinity may be ignored, and the usual theorems of classical and stratified Morse theory may be applied.

82, TITLE:      On Consequentialism and Fairness
http://arxiv.org/abs/2001.00329
AUTHORS:        Dallas Card ; Noah A. Smith
COMMENTS:       Updating to published version
HIGHLIGHT:      In this paper we provide a consequentialist critique of common definitions of fairness within machine learning, as well as a machine learning perspective on consequentialism.

83, TITLE:      Constant-Time Foundations for the New Spectre Era
http://arxiv.org/abs/1910.01755
AUTHORS:        Sunjay Cauligi ; Craig Disselkoen ; Klaus v. Gleissenthall ; Dean Tullsen ; Deian Stefan ; Tamara Rezk ; Gilles Barthe
COMMENTS:       To appear at PLDI '20
HIGHLIGHT:      We present an operational semantics and a formal definition of constant-time programs in this extended setting.

84, TITLE:      Conflict complexity is lower bounded by block sensitivity
http://arxiv.org/abs/1810.08873
AUTHORS:        Yaqiao Li
COMMENTS:       a discussion section added
HIGHLIGHT:      We propose to compare conflict complexity with certificate complexity, and explain why it could be interesting.

85, TITLE:      Interpretation and Simplification of Deep Forest
http://arxiv.org/abs/2001.04721
AUTHORS:        Sangwon Kim ; Mira Jeong ; Byoung Chul Ko
HIGHLIGHT:      This paper proposes a new method for interpreting and simplifying a black box model of a deep random forest (RF) using a proposed rule elimination.

86, TITLE:      Whence the Expected Free Energy?
http://arxiv.org/abs/2004.08128
AUTHORS:        Beren Millidge ; Alexander Tschantz ; Christopher L Buckley
COMMENTS:       24 pages, 0 figures. Reuploaded to correct typos in the original
HIGHLIGHT:      In this paper, we investigate the origins of the EFE in detail and show that it is not simply "the free energy in the future".

87, TITLE:      Optimal Covid-19 Pool Testing with a priori Information
http://arxiv.org/abs/2005.02940
AUTHORS:        Marc Beunardeau ; Éric Brier ; Noémie Cartier ; Aisling Connolly ; Nathanaël Courant ; Rémi Géraud-Stewart ; David Naccache ; Ofer Yifrach-Stav
HIGHLIGHT:      This paper describes how to optimally detect infected patients in pools, i.e. using a minimal number of tests to precisely identify them, given the a priori probabilities that each of the patients is healthy.

88, TITLE:      A Study of Non-autoregressive Model for Sequence Generation
http://arxiv.org/abs/2004.10454
AUTHORS:        Yi Ren ; Jinglin Liu ; Xu Tan ; Zhou Zhao ; Sheng Zhao ; Tie-Yan Liu
COMMENTS:       Accepted by ACL 2020
HIGHLIGHT:      To quantify such dependency, we propose an analysis model called CoMMA to characterize the difficulty of different NAR sequence generation tasks.

89, TITLE:      DIET: Lightweight Language Understanding for Dialogue Systems
http://arxiv.org/abs/2004.09936
AUTHORS:        Tanja Bunk ; Daksh Varshneya ; Vladimir Vlasov ; Alan Nichol
COMMENTS:       v3: Updated results for the best model
HIGHLIGHT:      We introduce the Dual Intent and Entity Transformer (DIET) architecture, and study the effectiveness of different pre-trained representations on intent and entity prediction, two common dialogue language understanding tasks.

90, TITLE:      Multi-scale fully convolutional neural networks for histopathology image segmentation: from nuclear aberrations to the global tissue architecture
http://arxiv.org/abs/1909.10726
AUTHORS:        Rüdiger Schmitz ; Frederic Madesta ; Maximilian Nielsen ; Jenny Krause ; René Werner ; Thomas Rösch
HIGHLIGHT:      To explicitly mimic how human pathologists combine multi-scale information, we introduce a family of multi-encoder fully-convolutional neural networks with deep fusion.

91, 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.

92, TITLE:      Image Matching across Wide Baselines: From Paper to Practice
http://arxiv.org/abs/2003.01587
AUTHORS:        Yuhe Jin ; Dmytro Mishkin ; Anastasiia Mishchuk ; Jiri Matas ; Pascal Fua ; Kwang Moo Yi ; Eduard Trulls
COMMENTS:       24 pages, 30 figures. Added experiments showing ground truth generation robustness
HIGHLIGHT:      We introduce a comprehensive benchmark for local features and robust estimation algorithms, focusing on the downstream task -- the accuracy of the reconstructed camera pose -- as our primary metric.

93, TITLE:      To Test Machine Comprehension, Start by Defining Comprehension
http://arxiv.org/abs/2005.01525
AUTHORS:        Jesse Dunietz ; Gregory Burnham ; Akash Bharadwaj ; Owen Rambow ; Jennifer Chu-Carroll ; David Ferrucci
COMMENTS:       Camera-ready ACL 2020 paper (Theme track). 9 pages; 3 figures; 1 table
HIGHLIGHT:      In this paper we make two key contributions.

94, TITLE:      Predicting Sharp and Accurate Occlusion Boundaries in Monocular Depth Estimation Using Displacement Fields
http://arxiv.org/abs/2002.12730
AUTHORS:        Michael Ramamonjisoa ; Yuming Du ; Vincent Lepetit
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      We instead learn to predict, given a depth map predicted by some reconstruction method, a 2D displacement field able to re-sample pixels around the occlusion boundaries into sharper reconstructions.

95, TITLE:      The $(1+(λ,λ))$ Genetic Algorithm for Permutations
http://arxiv.org/abs/2004.08664
AUTHORS:        Anton Bassin ; Maxim Buzdalov
COMMENTS:       This contribution is a slightly extended version of the paper accepted to the GECCO 2020 workshop on permutation-based problems
HIGHLIGHT:      We aim at improving this situation by proposing an adaptation of the $(1+(\lambda,\lambda))$ genetic algorithm to permutation-based problems.

96, TITLE:      Entity Synonym Discovery via Multipiece Bilateral Context Matching
http://arxiv.org/abs/1901.00056
AUTHORS:        Chenwei Zhang ; Yaliang Li ; Nan Du ; Wei Fan ; Philip S. Yu
COMMENTS:       In IJCAI 2020 as a long paper. Code and data are available at https://github.com/czhang99/SynonymNet
HIGHLIGHT:      To leverage diverse contexts where entities are mentioned, in this paper, we generalize the distributional hypothesis to a multi-context setting and propose a synonym discovery framework that detects entity synonyms from free-text corpora with considerations on effectiveness and robustness.

97, TITLE:      Distantly-Supervised Neural Relation Extraction with Side Information using BERT
http://arxiv.org/abs/2004.14443
AUTHORS:        Johny Moreira ; Chaina Oliveira ; David Macêdo ; Cleber Zanchettin ; Luciano Barbosa
HIGHLIGHT:      Considering that this method outperformed state-of-the-art baselines, in this paper, we propose a related approach to RESIDE also using additional side information, but simplifying the sentence encoding with BERT embeddings.

98, TITLE:      Intra-Ensemble in Neural Networks
http://arxiv.org/abs/1904.04466
AUTHORS:        Yuan Gao ; Zixiang Cai ; Lei Yu
HIGHLIGHT:      In this work, we propose Intra-Ensemble, an end-to-end ensemble strategy with stochastic channel recombination operations to train several sub-networks simultaneously within one neural network.

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

100, TITLE:     Is Sampling Heuristics Necessary in Training Deep Object Detectors?
http://arxiv.org/abs/1909.04868
AUTHORS:        Joya Chen ; Dong Liu ; Tong Xu ; Shiwei Wu ; Shilong Zhang ; Bin Luo ; Xuezheng Peng ; Enhong Chen
COMMENTS:       Tech Report (8 pages). Appreciate gratefully for your advice, please feel free to contact me: chenjoya@mail . ustc . edu . cn
HIGHLIGHT:      In this paper, we challenge the necessity of such hard/soft sampling heuristics in training deep object detectors.

101, TITLE:     Multiscale Collaborative Deep Models for Neural Machine Translation
http://arxiv.org/abs/2004.14021
AUTHORS:        Xiangpeng Wei ; Heng Yu ; Yue Hu ; Yue Zhang ; Rongxiang Weng ; Weihua Luo
COMMENTS:       ACL 2020
HIGHLIGHT:      In this paper, we present a MultiScale Collaborative (MSC) framework to ease the training of NMT models that are substantially deeper than those used previously.

102, TITLE:     Tuning Algorithms and Generators for Efficient Edge Inference
http://arxiv.org/abs/1908.02239
AUTHORS:        Rawan Naous ; Lazar Supic ; Yoonhwan Kang ; Ranko Sredojevic ; Anish Singhani ; Vladimir Stojanovic
HIGHLIGHT:      In this paper, we address these issues by creating a cross-layer software-hardware design framework that encompasses network training and model compression that is aware of and tuned to the underlying hardware architecture.

103, TITLE:     Improving BPSO-based feature selection applied to offline WI handwritten signature verification through overfitting control
http://arxiv.org/abs/2004.03373
AUTHORS:        Victor L. F. Souza ; Adriano L. I. Oliveira ; Rafael M. O. Cruz ; Robert Sabourin
HIGHLIGHT:      This paper investigates the presence of overfitting when using Binary Particle Swarm Optimization (BPSO) to perform the feature selection in a context of Handwritten Signature Verification (HSV).