2020.07.01.txt

==========New Papers==========
1, TITLE:       BitMix: Data Augmentation for Image Steganalysis
http://arxiv.org/abs/2006.16625
AUTHORS:        In-Jae Yu ; Wonhyuk Ahn ; Seung-Hun Nam ; Heung-Kyu Lee
HIGHLIGHT:      To overcome this limitation, we propose BitMix, a data augmentation method for spatial image steganalysis.

2, TITLE:       Boosting Deep Neural Networks with Geometrical Prior Knowledge: A Survey
http://arxiv.org/abs/2006.16867
AUTHORS:        Matthias Rath ; Alexandru Paul Condurache
COMMENTS:       Survey Paper
HIGHLIGHT:      In this survey, we try to give a concise overview about different approaches to incorporate geometrical prior knowledge into DNNs.

3, TITLE:       OccInpFlow: Occlusion-Inpainting Optical Flow Estimation by Unsupervised Learning
http://arxiv.org/abs/2006.16637
AUTHORS:        Kunming Luo ; Chuan Wang ; Nianjin Ye ; Shuaicheng Liu ; Jue Wang
HIGHLIGHT:      In this paper, we present OccInpFlow, an occlusion-inpainting framework to make full use of occlusion regions.

4, TITLE:       Counting Homomorphisms to $K_4$-minor-free Graphs, modulo 2
http://arxiv.org/abs/2006.16632
AUTHORS:        Jacob Focke ; Leslie Ann Goldberg ; Marc Roth ; Stanislav Živný
HIGHLIGHT:      Our proofs introduce a novel method of deriving hardness from globally defined substructures of the fixed graph $H$.

5, TITLE:       Model-Targeted Poisoning Attacks: Provable Convergence and Certified Bounds
http://arxiv.org/abs/2006.16469
AUTHORS:        Fnu Suya ; Saeed Mahloujifar ; David Evans ; Yuan Tian
COMMENTS:       21 pages, code available at: https://github.com/suyeecav/model-targeted-poisoning
HIGHLIGHT:      In this paper, we consider poisoning attacks where there is an adversary who has a particular target classifier in mind and hopes to induce a classifier close to that target by adding as few poisoning points as possible.

6, TITLE:       Learning to Read through Machine Teaching
http://arxiv.org/abs/2006.16470
AUTHORS:        Ayon Sen ; Christopher R. Cox ; Matthew Cooper Borkenhagen ; Mark S. Seidenberg ; Xiaojin Zhu
HIGHLIGHT:      We used a cognitively interesting neural network architecture to examine whether the sequence of learning trials could be structured to facilitate learning.

7, TITLE:       Primary Tumor Origin Classification of Lung Nodules in Spectral CT using Transfer Learning
http://arxiv.org/abs/2006.16633
AUTHORS:        Linde S. Hesse ; Pim A. de Jong ; Josien P. W. Pluim ; Veronika Cheplygina
COMMENTS:       MSc thesis Linde Hesse
HIGHLIGHT:      Therefore, we propose a detection and classification system for lung nodules in CT scans.

8, TITLE:       Object Detection under Rainy Conditions for Autonomous Vehicles
http://arxiv.org/abs/2006.16471
AUTHORS:        Mazin Hnewa ; Hayder Radha
HIGHLIGHT:      The main objective of this paper is to present a tutorial on state-of-the-art and emerging techniques that represent leading candidates for mitigating the influence of rainy conditions on an autonomous vehicle\textsc{'}s ability to detect objects.

9, TITLE:       Deep Learning Based Anticipatory Multi-Objective Eco-Routing Strategies for Connected and Automated Vehicles
http://arxiv.org/abs/2006.16472
AUTHORS:        Lama Alfaseeh ; Bilal Farooq
HIGHLIGHT:      This study exploits the advancements in information and communication technology (ICT), connected and automated vehicles (CAVs), and sensing, to develop anticipatory multi-objective eco-routing strategies.

10, TITLE:      Ontology-guided Semantic Composition for Zero-Shot Learning
http://arxiv.org/abs/2006.16917
AUTHORS:        Jiaoyan Chen ; Freddy Lecue ; Yuxia Geng ; Jeff Z. Pan ; Huajun Chen
COMMENTS:       Accepted by KR 2020 - 17th International Conference on Principles of Knowledge Representation and Reasoning
HIGHLIGHT:      Ontology-guided Semantic Composition for Zero-Shot Learning

11, TITLE:      MSNet: A Multilevel Instance Segmentation Network for Natural Disaster Damage Assessment in Aerial Videos
http://arxiv.org/abs/2006.16479
AUTHORS:        Xiaoyu Zhu ; Junwei Liang ; Alexander Hauptmann
HIGHLIGHT:      In this paper, we study the problem of efficiently assessing building damage after natural disasters like hurricanes, floods or fires, through aerial video analysis. We will release our data, models and code.

12, TITLE:      A Data-driven Neural Network Architecture for Sentiment Analysis
http://arxiv.org/abs/2006.16642
AUTHORS:        Erion Çano ; Maurizio Morisio
COMMENTS:       18 pages, 4 tables, 7 figures
HIGHLIGHT:      In this paper we present the creation steps of two big datasets of song emotions.

13, TITLE:      Vehicle Attribute Recognition by Appearance: Computer Vision Methods for Vehicle Type, Make and Model Classification
http://arxiv.org/abs/2006.16400
AUTHORS:        Xingyang Ni ; Heikki Huttunen
COMMENTS:       Published in Journal of Signal Processing Systems
HIGHLIGHT:      We survey a number of algorithms that identify vehicle properties ranging from coarse-grained level (vehicle type) to fine-grained level (vehicle make and model).

14, TITLE:      ANA at SemEval-2020 Task 4: mUlti-task learNIng for cOmmonsense reasoNing (UNION)
http://arxiv.org/abs/2006.16403
AUTHORS:        Anandh Perumal ; Chenyang Huang ; Amine Trabelsi ; Osmar R. Zaïane
COMMENTS:       7 pages, 1 figure, 3 tables, SemEval 2020
HIGHLIGHT:      In this paper, we describe our mUlti-task learNIng for cOmmonsense reasoNing (UNION) system submitted for Task C of the SemEval2020 Task 4, which is to generate a reason explaining why a given false statement is non-sensical.

15, TITLE:      Reading Between the Demographic Lines: Resolving Sources of Bias in Toxicity Classifiers
http://arxiv.org/abs/2006.16402
AUTHORS:        Elizabeth Reichert ; Helen Qiu ; Jasmine Bayrooti
COMMENTS:       8 pages, 13 figures
HIGHLIGHT:      Perspective API, a creation of Google technology incubator Jigsaw, is perhaps the most widely used toxicity classifier in industry; the model is employed by several online communities including The New York Times to identify and filter out toxic comments with the goal of preserving online safety.

16, TITLE:      Rethinking CNN-Based Pansharpening: Guided Colorization of Panchromatic Images via GANs
http://arxiv.org/abs/2006.16644
AUTHORS:        Furkan Ozcelik ; Ugur Alganci ; Elif Sertel ; Gozde Unal
COMMENTS:       F. Ozcelik, U. Alganci, E. Sertel, G. Unal, "Rethinking CNN-Based Pansharpening: Guided Colorization of Panchromatic Images via GANs", IEEE Transactions on Geoscience and Remote Sensing (TGRS), in press, 2020
HIGHLIGHT:      To that end, we propose a new self-supervised learning framework, where we treat pansharpening as a colorization problem, which brings an entirely novel perspective and solution to the problem compared to existing methods that base their solution solely on producing a super-resolution version of the multispectral image.

17, TITLE:      Accelerating Reinforcement Learning Agent with EEG-based Implicit Human Feedback
http://arxiv.org/abs/2006.16498
AUTHORS:        Duo Xu ; Mohit Agarwal ; Faramarz Fekri ; Raghupathy Sivakumar
HIGHLIGHT:      In this work, we investigate capturing human's intrinsic reactions as implicit (and natural) feedback through EEG in the form of error-related potentials (ErrP), providing a natural and direct way for humans to improve the RL agent learning.

18, TITLE:      Multi-sensory Integration in a Quantum-Like Robot Perception Model
http://arxiv.org/abs/2006.16404
AUTHORS:        Davide Lanza ; Paolo Solinas ; Fulvio Mastrogiovanni
COMMENTS:       Paper submitted to the 17th International Symposium on Experimental Robotics, Malta, Nov. 9-12, 2020
HIGHLIGHT:      In this paper, we generalize such a model for multi-sensory inputs, creating a multidimensional world representation directly based on sensor readings.

19, TITLE:      FVV Live: Real-Time, Low-Cost, Free Viewpoint Video
http://arxiv.org/abs/2006.16893
AUTHORS:        Daniel Berjón ; Pablo Carballeira ; Julián Cabrera ; Carlos Carmona ; Daniel Corregidor ; César Díaz ; Francisco Morán ; Narciso García
HIGHLIGHT:      FVV Live: Real-Time, Low-Cost, Free Viewpoint Video

20, TITLE:      Correction of Faulty Background Knowledge based on Condition Aware and Revise Transformer for Question Answering
http://arxiv.org/abs/2006.16722
AUTHORS:        Xinyan Zhao ; Xiao Feng ; Haoming Zhong ; Jun Yao ; Huanhuan Chen
HIGHLIGHT:      To alleviate the effect of defective condition values, this paper proposes condition aware and revise Transformer (CAR-Transformer).

21, TITLE:      Neural Datalog Through Time: Informed Temporal Modeling via Logical Specification
http://arxiv.org/abs/2006.16723
AUTHORS:        Hongyuan Mei ; Guanghui Qin ; Minjie Xu ; Jason Eisner
COMMENTS:       ICML 2020 (near-camera-ready version)
HIGHLIGHT:      To exploit domain-specific knowledge of how past events might affect an event's present probability, we propose using a temporal deductive database to track structured facts over time.

22, TITLE:      Improving robustness against common corruptions by covariate shift adaptation
http://arxiv.org/abs/2006.16971
AUTHORS:        Steffen Schneider ; Evgenia Rusak ; Luisa Eck ; Oliver Bringmann ; Wieland Brendel ; Matthias Bethge
HIGHLIGHT:      With the more robust AugMix model, we improve the state of the art from 56.5% mCE to 51.0% mCE.

23, TITLE:      Beyond accuracy: quantifying trial-by-trial behaviour of CNNs and humans by measuring error consistency
http://arxiv.org/abs/2006.16736
AUTHORS:        Robert Geirhos ; Kristof Meding ; Felix A. Wichmann
HIGHLIGHT:      Here we introduce trial-by-trial error consistency, a quantitative analysis for measuring whether two decision making systems systematically make errors on the same inputs.

24, TITLE:      Enabling Continual Learning with Differentiable Hebbian Plasticity
http://arxiv.org/abs/2006.16558
AUTHORS:        Vithursan Thangarasa ; Thomas Miconi ; Graham W. Taylor
COMMENTS:       Published as a conference paper at IJCNN 2020
HIGHLIGHT:      We propose a Differentiable Hebbian Consolidation model which is composed of a Differentiable Hebbian Plasticity (DHP) Softmax layer that adds a rapid learning plastic component (compressed episodic memory) to the fixed (slow changing) parameters of the softmax output layer; enabling learned representations to be retained for a longer timescale.

25, TITLE:      Towards Robust LiDAR-based Perception in Autonomous Driving: General Black-box Adversarial Sensor Attack and Countermeasures
http://arxiv.org/abs/2006.16974
AUTHORS:        Jiachen Sun ; Yulong Cao ; Qi Alfred Chen ; Z. Morley Mao
COMMENTS:       18 pages, 27 figures, to be published in USENIX Security 2020
HIGHLIGHT:      In this work, we perform the first study to explore the general vulnerability of current LiDAR-based perception architectures and discover that the ignored occlusion patterns in LiDAR point clouds make self-driving cars vulnerable to spoofing attacks.

26, TITLE:      Self-Supervised Learning of a Biologically-Inspired Visual Texture Model
http://arxiv.org/abs/2006.16976
AUTHORS:        Nikhil Parthasarathy ; Eero P. Simoncelli
COMMENTS:       15 pages, 7 figures
HIGHLIGHT:      We develop a model for representing visual texture in a low-dimensional feature space, along with a novel self-supervised learning objective that is used to train it on an unlabeled database of texture images.

27, TITLE:      Learning Post-Hoc Causal Explanations for Recommendation
http://arxiv.org/abs/2006.16977
AUTHORS:        Shuyuan Xu ; Yunqi Li ; Shuchang Liu ; Zuohui Fu ; Yongfeng Zhang
HIGHLIGHT:      In this paper, we propose to extract causal rules from the user interaction history as post-hoc explanations for the black-box sequential recommendation mechanisms, whilst maintain the predictive accuracy of the recommendation model.

28, TITLE:      Scaling Symbolic Methods using Gradients for Neural Model Explanation
http://arxiv.org/abs/2006.16322
AUTHORS:        Subham Sekhar Sahoo ; Subhashini Venugopalan ; Li Li ; Rishabh Singh ; Patrick Riley
HIGHLIGHT:      In this work, we propose a technique for combining gradient-based methods with symbolic techniques to scale such analyses and demonstrate its application for model explanation.

29, TITLE:      Learning to Combine Top-Down and Bottom-Up Signals in Recurrent Neural Networks with Attention over Modules
http://arxiv.org/abs/2006.16981
AUTHORS:        Sarthak Mittal ; Alex Lamb ; Anirudh Goyal ; Vikram Voleti ; Murray Shanahan ; Guillaume Lajoie ; Michael Mozer ; Yoshua Bengio
COMMENTS:       ICML 2020
HIGHLIGHT:      We explore deep recurrent neural net architectures in which bottom-up and top-down signals are dynamically combined using attention.

30, TITLE:      Universal linguistic inductive biases via meta-learning
http://arxiv.org/abs/2006.16324
AUTHORS:        R. Thomas McCoy ; Erin Grant ; Paul Smolensky ; Thomas L. Griffiths ; Tal Linzen
COMMENTS:       To appear in the Proceedings of the 42nd Annual Conference of the Cognitive Science Society
HIGHLIGHT:      We demonstrate this framework with a case study based on syllable structure.

31, TITLE:      Image-level Harmonization of Multi-Site Data using Image-and-Spatial Transformer Networks
http://arxiv.org/abs/2006.16741
AUTHORS:        R. Robinson ; Q. Dou ; D. C. Castro ; K. Kamnitsas ; M. de Groot ; R. M. Summers ; D. Rueckert ; B. Glocker
COMMENTS:       Accepted at MICCAI 2020
HIGHLIGHT:      We investigate the use of image-and-spatial transformer networks (ISTNs) to tackle domain shift in multi-site medical imaging data.

32, TITLE:      Method for the generation of depth images for view-based shape retrieval of 3D CAD model from partial point cloud
http://arxiv.org/abs/2006.16500
AUTHORS:        Hyungki Kim ; Moohyun Cha ; Duhwan Mun
COMMENTS:       39 pages
HIGHLIGHT:      In this paper, we propose a method of viewpoint and image resolution estimation method for view-based 3D shape retrieval from point cloud query.

33, TITLE:      Policy Gradient Optimization of Thompson Sampling Policies
http://arxiv.org/abs/2006.16507
AUTHORS:        Seungki Min ; Ciamac C. Moallemi ; Daniel J. Russo
HIGHLIGHT:      Our central insight is to view the posterior parameter sampled by Thompson sampling as a kind of pseudo-action.

34, TITLE:      Vehicle Re-ID for Surround-view Camera System
http://arxiv.org/abs/2006.16503
AUTHORS:        Zizhang Wu ; Man Wang ; Lingxiao Yin ; Weiwei Sun ; Jason Wang ; Huangbin Wu
COMMENTS:       CVPR 2020 workshop on Scalability in Autonomous Driving
HIGHLIGHT:      Thus, we present an integral vehicle Re-ID solution to address these problems. Besides, we will release the code and annotated fisheye dataset for the benefit of community.

35, TITLE:      On the Applicability of ML Fairness Notions
http://arxiv.org/abs/2006.16745
AUTHORS:        Karima Makhlouf ; Sami Zhioua ; Catuscia Palamidessi
HIGHLIGHT:      This paper is a survey of fairness notions that, unlike other surveys in the literature, addresses the question of "which notion of fairness is most suited to a given real-world scenario and why?"

36, TITLE:      Tackling Occlusion in Siamese Tracking with Structured Dropouts
http://arxiv.org/abs/2006.16571
AUTHORS:        Deepak K. Gupta ; Efstratios Gavves ; Arnold W. M. Smeulders
COMMENTS:       Accepted at ICPR2020
HIGHLIGHT:      In this paper, we propose a simple solution to simulate the effects of occlusion in the latent space.

37, TITLE:      Asymmetric metric learning for knowledge transfer
http://arxiv.org/abs/2006.16331
AUTHORS:        Mateusz Budnik ; Yannis Avrithis
HIGHLIGHT:      In this work, focusing on instance-level image retrieval, we study an asymmetric testing task, where the database is represented by the teacher and queries by the student.

38, TITLE:      Learning Sparse Prototypes for Text Generation
http://arxiv.org/abs/2006.16336
AUTHORS:        Junxian He ; Taylor Berg-Kirkpatrick ; Graham Neubig
COMMENTS:       8 pages of main content
HIGHLIGHT:      In this paper, we propose a novel generative model that automatically learns a \emph{sparse} prototype support set that, nonetheless, achieves strong language modeling performance.

39, TITLE:      Deep Isometric Learning for Visual Recognition
http://arxiv.org/abs/2006.16992
AUTHORS:        Haozhi Qi ; Chong You ; Xiaolong Wang ; Yi Ma ; Jitendra Malik
COMMENTS:       ICML 2020; Code: https://github.com/HaozhiQi/ISONet
HIGHLIGHT:      This paper shows that deep vanilla ConvNets without normalization nor skip connections can also be trained to achieve surprisingly good performance on standard image recognition benchmarks.

40, TITLE:      Material Recognition for Automated Progress Monitoring using Deep Learning Methods
http://arxiv.org/abs/2006.16344
AUTHORS:        Navid Ghassemi ; Hadi Mahami ; Mohammad Tayarani Darbandi ; Afshin Shoeibi ; Sadiq Hussain ; Farnad Nasirzadeh ; Roohallah Alizadehsani ; Darius Nahavandi ; Abbas Khosravi ; Saeid Nahavandi
HIGHLIGHT:      In this paper, we have addressed these issues and reached a state of the art performance, i.e., 97.35% accuracy rate for this task.

41, TITLE:      Early Exit Or Not: Resource-Efficient Blind Quality Enhancement for Compressed Images
http://arxiv.org/abs/2006.16581
AUTHORS:        Qunliang Xing ; Mai Xu ; Tianyi Li ; Zhenyu Guan
HIGHLIGHT:      In this paper, we propose a resource-efficient blind quality enhancement (RBQE) approach for compressed images.

42, TITLE:      Handling Missing Data in Decision Trees: A Probabilistic Approach
http://arxiv.org/abs/2006.16341
AUTHORS:        Pasha Khosravi ; Antonio Vergari ; YooJung Choi ; Yitao Liang ; Guy Van den Broeck
HIGHLIGHT:      In this paper, we tackle this problem by taking a probabilistic approach.

43, TITLE:      On the Demystification of Knowledge Distillation: A Residual Network Perspective
http://arxiv.org/abs/2006.16589
AUTHORS:        Nandan Kumar Jha ; Rajat Saini ; Sparsh Mittal
HIGHLIGHT:      We propose a novel interpretation of KD with the Trainee-Mentor hypothesis, which provides a holistic view of KD.

44, TITLE:      Terahertz Pulse Shaping Using Diffractive Legos
http://arxiv.org/abs/2006.16599
AUTHORS:        Muhammed Veli ; Deniz Mengu ; Nezih T. Yardimci ; Yi Luo ; Jingxi Li ; Yair Rivenson ; Mona Jarrahi ; Aydogan Ozcan
COMMENTS:       24 pages, 5 figures
HIGHLIGHT:      Here, we present a diffractive network, which is used to shape an arbitrary broadband pulse into a desired optical waveform, forming a compact pulse engineering system.

45, TITLE:      Technical Report: Auxiliary Tuning and its Application to Conditional Text Generation
http://arxiv.org/abs/2006.16823
AUTHORS:        Yoel Zeldes ; Dan Padnos ; Or Sharir ; Barak Peleg
HIGHLIGHT:      We introduce a simple and efficient method, called Auxiliary Tuning, for adapting a pre-trained Language Model to a novel task; we demonstrate this approach on the task of conditional text generation.

46, TITLE:      You Only Look Yourself: Unsupervised and Untrained Single Image Dehazing Neural Network
http://arxiv.org/abs/2006.16829
AUTHORS:        Boyun Li ; Yuanbiao Gou ; Shuhang Gu ; Jerry Zitao Liu ; Joey Tianyi Zhou ; Xi Peng
HIGHLIGHT:      In this paper, we study two challenging and less-touched problems in single image dehazing, namely, how to make deep learning achieve image dehazing without training on the ground-truth clean image (unsupervised) and a image collection (untrained).

47, TITLE:      Can Your Face Detector Do Anti-spoofing? Face Presentation Attack Detection with a Multi-Channel Face Detector
http://arxiv.org/abs/2006.16836
AUTHORS:        Anjith George ; Sebastien Marcel
COMMENTS:       9 pages,
HIGHLIGHT:      In this work, we try to reformulate the task of the face detector to detect real faces, thus eliminating the threat of presentation attacks.

48, TITLE:      SE3M: A Model for Software Effort Estimation Using Pre-trained Embedding Models
http://arxiv.org/abs/2006.16831
AUTHORS:        Eliane M. De Bortoli Fávero ; Dalcimar Casanova ; Andrey Ricardo Pimentel
HIGHLIGHT:      Estimating effort based on requirement texts presents many challenges, especially in obtaining viable features to infer effort.

49, TITLE:      A Framework for Learning Invariant Physical Relations in Multimodal Sensory Processing
http://arxiv.org/abs/2006.16607
AUTHORS:        Du Xiaorui ; Yavuzhan Erdem ; Immanuel Schweizer ; Cristian Axenie
HIGHLIGHT:      We describe the core system functionality when learning arbitrary non-linear relations in low-dimensional sensory data.

50, TITLE:      Symbolic Execution and Debugging Synchronization
http://arxiv.org/abs/2006.16601
AUTHORS:        Andrea Fioraldi
HIGHLIGHT:      In this thesis, we introduce the idea of combining symbolic execution with dynamic analysis for reverse engineering.

51, TITLE:      Mobile Link Prediction: Automated Creation and Crowd-sourced Validation of Knowledge Graphs
http://arxiv.org/abs/2006.16858
AUTHORS:        Mark Christopher Ballandies ; Evangelos Pournaras
HIGHLIGHT:      This paper introduces a novel pervasive knowledge graph builder that brings together automation, experts' and crowd-sourced citizens' knowledge.

52, TITLE:      Ideal Membership Problem for Boolean Minority
http://arxiv.org/abs/2006.16422
AUTHORS:        Arpitha P. Bharathi ; Monaldo Mastrolilli
COMMENTS:       21 pages
HIGHLIGHT:      In this paper we consider the IMP for polynomial ideals encoding combinatorial problems and where the input polynomial $f$ has degree at most $d=O(1)$ (we call this problem IMP$_d$).

53, TITLE:      Learning Patterns of Tourist Movement and Photography from Geotagged Photos at Archaeological Heritage Sites in Cuzco, Peru
http://arxiv.org/abs/2006.16424
AUTHORS:        Nicole D. Payntar ; Wei-Lin Hsiao ; R. Alan Covey ; Kristen Grauman
COMMENTS:       Accepted to Tourism Management
HIGHLIGHT:      Leveraging large-scale in-the-wild tourist photos, our goals are to (1) understand how the intensification of tourism intersects with heritage regulations and social media, aiding in the articulation of travel patterns across Cuzco's heritage landscape; and to (2) assess how aesthetic preferences and visuality become entangled with the rapidly evolving expectations of tourists, whose travel narratives are curated on social media and grounded in historic site representations.

54, TITLE:      GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding
http://arxiv.org/abs/2006.16668
AUTHORS:        Dmitry Lepikhin ; HyoukJoong Lee ; Yuanzhong Xu ; Dehao Chen ; Orhan Firat ; Yanping Huang ; Maxim Krikun ; Noam Shazeer ; Zhifeng Chen
HIGHLIGHT:      We demonstrate that such a giant model can efficiently be trained on 2048 TPU v3 accelerators in 4 days to achieve far superior quality for translation from 100 languages to English compared to the prior art.

55, TITLE:      EasyQuant: Post-training Quantization via Scale Optimization
http://arxiv.org/abs/2006.16669
AUTHORS:        Di Wu ; Qi Tang ; Yongle Zhao ; Ming Zhang ; Ying Fu ; Debing Zhang
HIGHLIGHT:      In this paper, we present an efficient and simple post-training method via scale optimization, named EasyQuant (EQ),that could obtain comparable accuracy with the training-based method.Specifically, we first alternately optimize scales of weights and activations for all layers target at convolutional outputs to further obtain the high quantization precision.

56, TITLE:      Biologically Inspired Mechanisms for Adversarial Robustness
http://arxiv.org/abs/2006.16427
AUTHORS:        Manish V. Reddy ; Andrzej Banburski ; Nishka Pant ; Tomaso Poggio
COMMENTS:       25 pages, 15 figures
HIGHLIGHT:      In this work, we investigate the role of two biologically plausible mechanisms in adversarial robustness.

57, TITLE:      PriorGAN: Real Data Prior for Generative Adversarial Nets
http://arxiv.org/abs/2006.16990
AUTHORS:        Shuyang Gu ; Jianmin Bao ; Dong Chen ; Fang Wen
HIGHLIGHT:      To address these two issues, we propose a novel prior that captures the whole real data distribution for GANs, which are called PriorGANs.

58, TITLE:      A Simple Domain Shifting Networkfor Generating Low Quality Images
http://arxiv.org/abs/2006.16621
AUTHORS:        Guruprasad Hegde ; Avinash Nittur Ramesh ; Kanchana Vaishnavi Gandikota ; Roman Obermaisser ; Michael Moeller
COMMENTS:       accepted ICPR 2020
HIGHLIGHT:      As a solution we propose to train a network on degrading the quality images in order to mimic specific low quality imaging systems.

59, TITLE:      Cross-Scale Internal Graph Neural Network for Image Super-Resolution
http://arxiv.org/abs/2006.16673
AUTHORS:        Shangchen Zhou ; Jiawei Zhang ; Wangmeng Zuo ; Chen Change Loy
HIGHLIGHT:      In this paper, we explore the cross-scale patch recurrence property of a natural image, i.e., similar patches tend to recur many times across different scales.

60, TITLE:      Predicting Sample Collision with Neural Networks
http://arxiv.org/abs/2006.16868
AUTHORS:        Tuan Tran ; Jory Denny ; Chinwe Ekenna
COMMENTS:       7 pages, 7 figures
HIGHLIGHT:      In this work, we present a framework to address the cost of expensive primitive operations in sampling-based motion planning.

61, TITLE:      On Finite Entailment of Non-Local Queries in Description Logics
http://arxiv.org/abs/2006.16869
AUTHORS:        Tomasz Gogacz ; Víctor Gutiérrez-Basulto ; Albert Gutowski ; Yazmín Ibáñez-García ; Filip Murlak
COMMENTS:       25 pages, 4 Figures, Accepted at KR 2020
HIGHLIGHT:      We study the problem of finite entailment of ontology-mediated queries.

62, TITLE:      Training highly effective connectivities within neural networks with randomly initialized, fixed weights
http://arxiv.org/abs/2006.16627
AUTHORS:        Cristian Ivan ; Razvan Florian
COMMENTS:       13 pages, 12 figures
HIGHLIGHT:      We present some novel, straightforward methods for training the connection graph of a randomly initialized neural network without training the weights.

63, TITLE:      Needle tip force estimation by deep learning from raw spectral OCT data
http://arxiv.org/abs/2006.16675
AUTHORS:        M. Gromniak ; N. Gessert ; T. Saathoff ; A. Schlaefer
COMMENTS:       Accepted for publication in IJCARS
HIGHLIGHT:      Here, we study how to calibrate OCT to sense forces, e.g. during robotic needle placement.

64, TITLE:      Efficient Continuous Pareto Exploration in Multi-Task Learning
http://arxiv.org/abs/2006.16434
AUTHORS:        Pingchuan Ma ; Tao Du ; Wojciech Matusik
COMMENTS:       ICML 2020 camera-ready. Code: https://github.com/mit-gfx/ContinuousParetoMTL
HIGHLIGHT:      We present a novel, efficient method that generates locally continuous Pareto sets and Pareto fronts, which opens up the possibility of continuous analysis of Pareto optimal solutions in machine learning problems.

65, TITLE:      Quantitative Evaluation of Endoscopic SLAM Methods: EndoSLAM Dataset
http://arxiv.org/abs/2006.16670
AUTHORS:        Kutsev Bengisu Ozyoruk ; Kagan Incetan ; Gulfize Coskun ; Guliz Irem Gokceler ; Yasin Almalioglu ; Faisal Mahmood ; Nicholas J. Durr ; Eva Curto ; Luis Perdigoto ; Marina Oliveira ; Helder Araujo ; Henrique Alexandrino ; Hunter B. Gilbert ; Mehmet Turan
COMMENTS:       27 pages, 16 figures
HIGHLIGHT:      With this paper, we introduce a comprehensive endoscopic SLAM dataset containing both capsule and standard endoscopy recordings.

66, TITLE:      Mitigating Manipulation in Peer Review via Randomized Reviewer Assignments
http://arxiv.org/abs/2006.16437
AUTHORS:        Steven Jecmen ; Hanrui Zhang ; Ryan Liu ; Nihar B. Shah ; Vincent Conitzer ; Fei Fang
COMMENTS:       27 pages, 9 figures
HIGHLIGHT:      On the conceptual front, we identify connections between these three problems and present a framework that brings all these challenges under a common umbrella.

67, TITLE:      R2-B2: Recursive Reasoning-Based Bayesian Optimization for No-Regret Learning in Games
http://arxiv.org/abs/2006.16679
AUTHORS:        Zhongxiang Dai ; Yizhou Chen ; Kian Hsiang Low ; Patrick Jaillet ; Teck-Hua Ho
COMMENTS:       Accepted to 37th International Conference on Machine Learning (ICML 2020), Extended version with proofs and additional experimental details and results, 27 pages
HIGHLIGHT:      This paper presents a recursive reasoning formalism of Bayesian optimization (BO) to model the reasoning process in the interactions between boundedly rational, self-interested agents with unknown, complex, and costly-to-evaluate payoff functions in repeated games, which we call Recursive Reasoning-Based BO (R2-B2).

68, TITLE:      Spiking Associative Memory for Spatio-Temporal Patterns
http://arxiv.org/abs/2006.16684
AUTHORS:        Simon Davidson ; Stephen B. Furber ; Oliver Rhodes
HIGHLIGHT:      This paper seeks to remedy this with two significant advances.

69, TITLE:      Maximum Entropy Models for Fast Adaptation
http://arxiv.org/abs/2006.16524
AUTHORS:        Samarth Sinha ; Anirudh Goyal ; Animesh Garg
HIGHLIGHT:      In this paper, we propose a regularization scheme using a max-entropy prior on the learned features of a neural network; such that the extracted features make minimal assumptions about the training data.

70, TITLE:      Actionable Attribution Maps for Scientific Machine Learning
http://arxiv.org/abs/2006.16533
AUTHORS:        Shusen Liu ; Bhavya Kailkhura ; Jize Zhang ; Anna M. Hiszpanski ; Emily Robertson ; Donald Loveland ; T. Yong-Jin Han
HIGHLIGHT:      In this work, we propose techniques for exploring the behavior of deep learning models by injecting domain-specific actionable concepts as tunable ``knobs'' in the analysis pipeline.

71, TITLE:      Multi-Partition Embedding Interaction with Block Term Format for Knowledge Graph Completion
http://arxiv.org/abs/2006.16365
AUTHORS:        Hung Nghiep Tran ; Atsuhiro Takasu
COMMENTS:       ECAI 2020. Including state-of-the-art results for very small models in appendix
HIGHLIGHT:      In this work, we propose the multi-partition embedding interaction (MEI) model with block term format to systematically address this problem.

72, TITLE:      Multi-Head Attention: Collaborate Instead of Concatenate
http://arxiv.org/abs/2006.16362
AUTHORS:        Jean-Baptiste Cordonnier ; Andreas Loukas ; Martin Jaggi
HIGHLIGHT:      Motivated by the observation that trained attention heads share common key/query projections, we propose a collaborative multi-head attention layer that enables heads to learn shared projections.

73, TITLE:      Classification of cancer pathology reports: a large-scale comparative study
http://arxiv.org/abs/2006.16370
AUTHORS:        Stefano Martina ; Leonardo Ventura ; Paolo Frasconi
COMMENTS:       10 pages, 6 figures, 3 tables, accepted for publication in IEEE Journal of Biomedical and Health Informatics (J-BHI)
HIGHLIGHT:      We present results on a large dataset (more than 80 000 labeled and 1 500 000 unlabeled anonymized reports written in Italian and collected from hospitals in Tuscany over more than a decade) and with a large number of classes (134 morphological classes and 61 topographical classes).

74, TITLE:      An EM Approach to Non-autoregressive Conditional Sequence Generation
http://arxiv.org/abs/2006.16378
AUTHORS:        Zhiqing Sun ; Yiming Yang
COMMENTS:       Accepted to ICML 2020
HIGHLIGHT:      This paper proposes a new approach that jointly optimizes both AR and NAR models in a unified Expectation-Maximization (EM) framework.

75, TITLE:      ERNIE-ViL: Knowledge Enhanced Vision-Language Representations Through Scene Graph
http://arxiv.org/abs/2006.16934
AUTHORS:        Fei Yu ; Jiji Tang ; Weichong Yin ; Yu Sun ; Hao Tian ; Hua Wu ; Haifeng Wang
HIGHLIGHT:      We propose a knowledge-enhanced approach, ERNIE-ViL, to learn joint representations of vision and language.

76, TITLE:      Complexity of Modification Problems for Best Match Graphs
http://arxiv.org/abs/2006.16909
AUTHORS:        David Schaller ; Peter F. Stadler ; Marc Hellmuth
HIGHLIGHT:      The arc modification problems for BMGs therefore provide a mean of correcting and improving the initial estimates of the best matches.

77, TITLE:      Classification Confidence Estimation with Test-Time Data-Augmentation
http://arxiv.org/abs/2006.16705
AUTHORS:        Yuval Bahat ; Gregory Shakhnarovich
HIGHLIGHT:      To this end, we propose a novel approach for classification confidence estimation.

78, TITLE:      A Survey on Recent Progress in the Theory of Evolutionary Algorithms for Discrete Optimization
http://arxiv.org/abs/2006.16709
AUTHORS:        Benjamin Doerr ; Frank Neumann
HIGHLIGHT:      The theory of evolutionary computation for discrete search spaces has made a lot of progress during the last ten years.

79, TITLE:      Large-scale inference of liver fat with neural networks on UK Biobank body MRI
http://arxiv.org/abs/2006.16777
AUTHORS:        Taro Langner ; Robin Strand ; Håkan Ahlström ; Joel Kullberg
HIGHLIGHT:      In this work, a novel framework for automated inference of liver fat from UK Biobank neck-to-knee body MRI is proposed.

80, TITLE:      Bounded Rationality in Las Vegas: Probabilistic Finite Automata PlayMulti-Armed Bandits
http://arxiv.org/abs/2006.16950
AUTHORS:        Xinming Liu ; Joseph Y. Halpern
COMMENTS:       10 pages, 4 pages, Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI), PMLR volume 124, 2020
HIGHLIGHT:      To test this hypothesis, we consider the multi-armed bandit (MAB) problem.

81, TITLE:      ITSELF: Iterative Saliency Estimation fLexible Framework
http://arxiv.org/abs/2006.16956
AUTHORS:        Leonardo de Melo Joao ; Felipe de Castro Belem ; Alexandre Xavier Falcao
HIGHLIGHT:      By combining a saliency-based superpixel algorithm to a superpixel-based saliency estimator, we propose a novel saliency/superpixel self-improving loop to iteratively enhance saliency maps.

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

83, TITLE:      Theory-Inspired Path-Regularized Differential Network Architecture Search
http://arxiv.org/abs/2006.16537
AUTHORS:        Pan Zhou ; Caiming Xiong ; Richard Socher ; Steven C. H. Hoi
HIGHLIGHT:      In this work, we solve this problem by theoretically analyzing the effects of various types of operations, e.g. convolution, skip connection and zero operation, to the network optimization.

84, TITLE:      Binary intersection formalized
http://arxiv.org/abs/2006.16711
AUTHORS:        Štěpán Holub ; Štěpán Starosta
HIGHLIGHT:      We provide a reformulation and a formalization of the classical result by Juhani Karhum\"aki characterizing intersections of two languages of the form $\{x,y\}^*\cap \{u,v\}^*$.

85, TITLE:      Model-based Reinforcement Learning: A Survey
http://arxiv.org/abs/2006.16712
AUTHORS:        Thomas M. Moerland ; Joost Broekens ; Catholijn M. Jonker
HIGHLIGHT:      This paper presents a survey of the integration of both fields, better known as model-based reinforcement learning.

86, TITLE:      A Compact Gated-Synapse Model for Neuromorphic Circuits
http://arxiv.org/abs/2006.16302
AUTHORS:        Alexander Jones ; Rashmi Jha
COMMENTS:       Submitted to IEEE Transactions on Computer-Aided Design for Integrated Circuits and Systems for review. "This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible."
HIGHLIGHT:      The behavioral theory of the model is described in detail along with a full list of the default parameter settings.

87, TITLE:      Adversarial Learning for Debiasing Knowledge Graph Embeddings
http://arxiv.org/abs/2006.16309
AUTHORS:        Mario Arduini ; Lorenzo Noci ; Federico Pirovano ; Ce Zhang ; Yash Raj Shrestha ; Bibek Paudel
HIGHLIGHT:      As a preliminary solution to debiasing KGs, we introduce a novel framework to filter out the sensitive attribute information from the KG embeddings, which we call FAN (Filtering Adversarial Network).

88, TITLE:      Hand-drawn Symbol Recognition of Surgical Flowsheet Graphs with Deep Image Segmentation
http://arxiv.org/abs/2006.16546
AUTHORS:        William Adorno III ; Angela Yi ; Marcel Durieux ; Donald Brown
COMMENTS:       8 pages, 4 figures, BioInformatics And BioEngineering 2020
HIGHLIGHT:      In this paper, we present a deep image segmentation approach using a U-Net architecture that can detect hand-drawn symbols on a flowsheet graph.

89, TITLE:      Leveraging Temporal Information for 3D Detection and Domain Adaptation
http://arxiv.org/abs/2006.16796
AUTHORS:        Cunjun Yu ; Zhongang Cai ; Daxuan Ren ; Haiyu Zhao
HIGHLIGHT:      In this report, we describe a simple way to pass such information in the learning pipeline by adding timestamps to the point clouds, which shows consistent improvements across all three classes.

90, TITLE:      On Bellman's Optimality Principle for zs-POSGs
http://arxiv.org/abs/2006.16395
AUTHORS:        Olivier Buffet ; Jilles Dibangoye ; Aurélien Delage ; Abdallah Saffidine ; Vincent Thomas
COMMENTS:       18 pages, 0 figures, 1 algorithm
HIGHLIGHT:      Here we show how it can apply to (infinite horizon) 2-player zero-sum partially observable stochastic games (zs-POSGs) by (i) taking a central planner's viewpoint, which can only reason on a sufficient statistic called occupancy state, and (ii) turning such problems into zero-sum occupancy Markov games (zs-OMGs).

91, TITLE:      Learning and Planning in Average-Reward Markov Decision Processes
http://arxiv.org/abs/2006.16318
AUTHORS:        Yi Wan ; Abhishek Naik ; Richard S. Sutton
HIGHLIGHT:      We introduce improved learning and planning algorithms for average-reward MDPs, including 1) the first general proven-convergent off-policy model-free control algorithm without reference states, 2) the first proven-convergent off-policy model-free prediction algorithm, and 3) the first learning algorithms that converge to the actual value function rather than to the value function plus an offset.



==========Updates to Previous Papers==========
1, TITLE:       Deep Reinforcement Learning Designed Shinnar-Le Roux RF Pulse using Root-Flipping: DeepRF_SLR
http://arxiv.org/abs/1912.09015
AUTHORS:        Dongmyung Shin ; Sooyeon Ji ; Doohee Lee ; Jieun Lee ; Se-Hong Oh ; Jongho Lee
HIGHLIGHT:      A novel approach of applying deep reinforcement learning to an RF pulse design is introduced.

2, TITLE:       Neuro-Symbolic Visual Reasoning: Disentangling "Visual" from "Reasoning"
http://arxiv.org/abs/2006.11524
AUTHORS:        Saeed Amizadeh ; Hamid Palangi ; Oleksandr Polozov ; Yichen Huang ; Kazuhito Koishida
COMMENTS:       To be published in Proceedings of the 37th International Conference on Machine Learning (ICML), Vienna, Austria, PMLR 119, 2020
HIGHLIGHT:      To address this, we propose (1) a framework to isolate and evaluate the reasoning aspect of VQA separately from its perception, and (2) a novel top-down calibration technique that allows the model to answer reasoning questions even with imperfect perception.

3, 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.

4, TITLE:       Global Wheat Head Detection (GWHD) dataset: a large and diverse dataset of high resolution RGB labelled images to develop and benchmark wheat head detection methods
http://arxiv.org/abs/2005.02162
AUTHORS:        E. David ; S. Madec ; P. Sadeghi-Tehran ; H. Aasen ; B. Zheng ; S. Liu ; N. Kirchgessner ; G. Ishikawa ; K. Nagasawa ; M. A. Badhon ; C. Pozniak ; B. de Solan ; A. Hund ; S. C. Chapman ; F. Baret ; I. Stavness ; W. Guo
COMMENTS:       16 pages, 7 figures, Dataset paper
HIGHLIGHT:      Several studies developed methods for wheat head detection from high-resolution RGB imagery. Through a joint international collaborative effort, we have built a large, diverse and well-labelled dataset, the Global Wheat Head detection (GWHD) dataset.

5, TITLE:       Non-Autoregressive Machine Translation with Disentangled Context Transformer
http://arxiv.org/abs/2001.05136
AUTHORS:        Jungo Kasai ; James Cross ; Marjan Ghazvininejad ; Jiatao Gu
COMMENTS:       ICML 2020
HIGHLIGHT:      We propose an attention-masking based model, called Disentangled Context (DisCo) transformer, that simultaneously generates all tokens given different contexts.

6, TITLE:       Neural Machine Translation with Noisy Lexical Constraints
http://arxiv.org/abs/1908.04664
AUTHORS:        Huayang Li ; Guoping Huang ; Lemao Liu
COMMENTS:       This paper was accepted by TASLP. See https://ieeexplore.ieee.org/document/9108255/ for the final version
HIGHLIGHT:      We present a novel framework that treats constraints as external memories.

7, TITLE:       A general approach to progressive learning
http://arxiv.org/abs/2004.12908
AUTHORS:        Joshua T. Vogelstein ; Hayden S. Helm ; Ronak D. Mehta ; Jayanta Dey ; Will LeVine ; Weiwei Yang ; Bryan Tower ; Jonathan Larson ; Chris White ; Carey E. Priebe
HIGHLIGHT:      We propose representation ensembling, as opposed to learner ensembling (e.g., bagging), to address progressive learning.

8, TITLE:       A Survey on Deep Learning for Localization and Mapping: Towards the Age of Spatial Machine Intelligence
http://arxiv.org/abs/2006.12567
AUTHORS:        Changhao Chen ; Bing Wang ; Chris Xiaoxuan Lu ; Niki Trigoni ; Andrew Markham
COMMENTS:       26 pages, 10 figures. Project website: https://github.com/changhao-chen/deep-learning-localization-mapping
HIGHLIGHT:      In this work, we provide a comprehensive survey, and propose a new taxonomy for localization and mapping using deep learning.

9, TITLE:       LP-SparseMAP: Differentiable Relaxed Optimization for Sparse Structured Prediction
http://arxiv.org/abs/2001.04437
AUTHORS:        Vlad Niculae ; André F. T. Martins
COMMENTS:       33 pages, 5 tables, 4 figures. ICML 2020
HIGHLIGHT:      In this paper, we introduce LP-SparseMAP, an extension of SparseMAP that addresses this limitation via a local polytope relaxation.

10, TITLE:      A Tandem Learning Rule for Effective Training and Rapid Inference of Deep Spiking Neural Networks
http://arxiv.org/abs/1907.01167
AUTHORS:        Jibin Wu ; Yansong Chua ; Malu Zhang ; Guoqi Li ; Haizhou Li ; Kay Chen Tan
HIGHLIGHT:      In this work, we propose a tandem learning framework, that consists of an SNN and an Artificial Neural Network (ANN) coupled through weight sharing.

11, 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.

12, TITLE:      Contestable Black Boxes
http://arxiv.org/abs/2006.05133
AUTHORS:        Andrea Aler Tubella ; Andreas Theodorou ; Virginia Dignum ; Loizos Michael
COMMENTS:       Accepted at RuleML 2020 as a short paper
HIGHLIGHT:      This paper investigates the type of assurances that are needed in the contesting process when algorithmic black-boxes are involved, opening new questions about the interplay of contestability and explainability.

13, TITLE:      SAR Image Despeckling by Deep Neural Networks: from a pre-trained model to an end-to-end training strategy
http://arxiv.org/abs/2006.15559
AUTHORS:        Emanuele Dalsasso ; Xiangli Yang ; Loïc Denis ; Florence Tupin ; Wen Yang
COMMENTS:       Notebook with Colab compatibility is available at https://github.com/emanueledalsasso/SAR-CNN
HIGHLIGHT:      To handle this problem, this paper analyzes different strategies one can adopt, depending on the speckle removal task one wishes to perform and the availability of multitemporal stacks of SAR data.

14, TITLE:      Reinforcement Learning for Robotic Manipulation using Simulated Locomotion Demonstrations
http://arxiv.org/abs/1910.07294
AUTHORS:        Ozsel Kilinc ; Yang Hu ; Giovanni Montana
HIGHLIGHT:      In this paper we propose a novel approach that introduces object-level demonstrations, i.e. examples of where the objects should be at any state.

15, TITLE:      Predicting Subjective Features from Questions on QA Websites using BERT
http://arxiv.org/abs/2002.10107
AUTHORS:        Issa Annamoradnejad ; Mohammadamin Fazli ; Jafar Habibi
COMMENTS:       5 pages, 4 figures, 2 tables
HIGHLIGHT:      Therefore, with the overall goal of providing solutions for automating moderation actions in Q&A websites, we aim to provide a model to predict 20 quality or subjective aspects of questions in QA websites.

16, TITLE:      The Partial Response Network: a neural network nomogram
http://arxiv.org/abs/1908.05978
AUTHORS:        Paulo J. G. Lisboa ; Sandra Ortega-Martorell ; Sadie Cashman ; Ivan Olier
COMMENTS:       21 pages, 8 figures
HIGHLIGHT:      A method is proposed to infer a Generalised Additive Neural Network (GANN) from an inital Multi-layer Perceptron (MLP).

17, TITLE:      Deep Molecular Programming: A Natural Implementation of Binary-Weight ReLU Neural Networks
http://arxiv.org/abs/2003.13720
AUTHORS:        Marko Vasic ; Cameron Chalk ; Sarfraz Khurshid ; David Soloveichik
HIGHLIGHT:      We discover a surprisingly tight connection between a popular class of neural networks (binary-weight ReLU aka BinaryConnect) and a class of coupled chemical reactions that are absolutely robust to reaction rates.

18, TITLE:      Super-resolution Variational Auto-Encoders
http://arxiv.org/abs/2006.05218
AUTHORS:        Ioannis Gatopoulos ; Maarten Stol ; Jakub M. Tomczak
COMMENTS:       13 pages, 11 figures, 3 tables. Code available at: https://github.com/ioangatop/srVAE
HIGHLIGHT:      Here, we propose to enhance VAEs by adding a random variable that is a downscaled version of the original image and still use the log-likelihood function as the learning objective.

19, TITLE:      AutoML-Zero: Evolving Machine Learning Algorithms From Scratch
http://arxiv.org/abs/2003.03384
AUTHORS:        Esteban Real ; Chen Liang ; David R. So ; Quoc V. Le
COMMENTS:       Accepted for publication at the 37th International Conference on Machine Learning (ICML 2020). Near camera-ready version
HIGHLIGHT:      Our goal is to show that AutoML can go further: it is possible today to automatically discover complete machine learning algorithms just using basic mathematical operations as building blocks.

20, TITLE:      Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels
http://arxiv.org/abs/1911.09781
AUTHORS:        Lu Jiang ; Di Huang ; Mason Liu ; Weilong Yang
COMMENTS:       published at ICML 2020
HIGHLIGHT:      This paper makes three contributions.

21, TITLE:      Automatic Shortcut Removal for Self-Supervised Representation Learning
http://arxiv.org/abs/2002.08822
AUTHORS:        Matthias Minderer ; Olivier Bachem ; Neil Houlsby ; Michael Tschannen
HIGHLIGHT:      Here, we propose a general framework for mitigating the effect shortcut features.

22, TITLE:      Problems with Shapley-value-based explanations as feature importance measures
http://arxiv.org/abs/2002.11097
AUTHORS:        I. Elizabeth Kumar ; Suresh Venkatasubramanian ; Carlos Scheidegger ; Sorelle Friedler
COMMENTS:       Accepted to ICML 2020
HIGHLIGHT:      Game-theoretic formulations of feature importance have become popular as a way to "explain" machine learning models.

23, TITLE:      Learning De-biased Representations with Biased Representations
http://arxiv.org/abs/1910.02806
AUTHORS:        Hyojin Bahng ; Sanghyuk Chun ; Sangdoo Yun ; Jaegul Choo ; Seong Joon Oh
COMMENTS:       Accepted to ICML 2020. Code available at https://github.com/clovaai/rebias
HIGHLIGHT:      In this work, we propose a novel framework to train a de-biased representation by encouraging it to be different from a set of representations that are biased by design.

24, TITLE:      Bidirectional Generative Modeling Using Adversarial Gradient Estimation
http://arxiv.org/abs/2002.09161
AUTHORS:        Xinwei Shen ; Tong Zhang ; Kani Chen
HIGHLIGHT:      We present a new optimization method for this formulation, where the gradient is computed using an adversarially learned discriminator.

25, TITLE:      Teach Me What You Want to Play: Learning Variants of Connect Four through Human-Robot Interaction
http://arxiv.org/abs/2001.01004
AUTHORS:        Ali Ayub ; Alan R. Wagner
COMMENTS:       Under Review at ICSR2020
HIGHLIGHT:      We combine aspects of learning by demonstration, active learning, and game theory allowing a robot to leverage its developing representation of the game to conduct question/answer sessions with a person, thus filling in gaps in its knowledge.

26, TITLE:      Decoupled Certainty-Driven Consistency Loss for Semi-supervised Learning
http://arxiv.org/abs/1901.05657
AUTHORS:        Lu Liu ; Yiting Li ; Robby T. Tan
HIGHLIGHT:      In this paper, we propose a novel certainty-driven consistency loss (CCL) that exploits the predictive uncertainty information in the consistency loss to let the student dynamically learn from reliable targets.

27, TITLE:      Attentive Group Equivariant Convolutional Networks
http://arxiv.org/abs/2002.03830
AUTHORS:        David W. Romero ; Erik J. Bekkers ; Jakub M. Tomczak ; Mark Hoogendoorn
COMMENTS:       Proceedings of the 37th International Conference on Machine Learning (ICML), 2020
HIGHLIGHT:      In this paper, we present attentive group equivariant convolutions, a generalization of the group convolution, in which attention is applied during the course of convolution to accentuate meaningful symmetry combinations and suppress non-plausible, misleading ones.

28, TITLE:      Confidence-Calibrated Adversarial Training: Generalizing to Unseen Attacks
http://arxiv.org/abs/1910.06259
AUTHORS:        David Stutz ; Matthias Hein ; Bernt Schiele
HIGHLIGHT:      For each threat model, we use $7$ attacks with up to $50$ restarts and $5000$ iterations and report worst-case robust test error, extended to our confidence-thresholded setting, across all attacks.

29, TITLE:      Data preprocessing to mitigate bias: A maximum entropy based approach
http://arxiv.org/abs/1906.02164
AUTHORS:        L. Elisa Celis ; Vijay Keswani ; Nisheeth K. Vishnoi
HIGHLIGHT:      This paper presents an algorithmic framework that can be used as a data preprocessing method towards mitigating such bias.

30, TITLE:      Convolutional Tensor-Train LSTM for Spatio-temporal Learning
http://arxiv.org/abs/2002.09131
AUTHORS:        Jiahao Su ; Wonmin Byeon ; Jean Kossaifi ; Furong Huang ; Jan Kautz ; Animashree Anandkumar
COMMENTS:       Jiahao Su and Wonmin Byeon contributed equally to this work. 20 pages, 14 figures
HIGHLIGHT:      In this paper, we propose a higher-order convolutional LSTM model that can efficiently learn these correlations, along with a succinct representations of the history.

31, TITLE:      You Only Watch Once: A Unified CNN Architecture for Real-Time Spatiotemporal Action Localization
http://arxiv.org/abs/1911.06644
AUTHORS:        Okan Köpüklü ; Xiangyu Wei ; Gerhard Rigoll
HIGHLIGHT:      In this work, we present YOWO, a unified CNN architecture for real-time spatiotemporal action localization in video streams.

32, TITLE:      The Origins and Prevalence of Texture Bias in Convolutional Neural Networks
http://arxiv.org/abs/1911.09071
AUTHORS:        Katherine L. Hermann ; Ting Chen ; Simon Kornblith
HIGHLIGHT:      By taking less aggressive random crops at training time and applying simple, naturalistic augmentation (color distortion, noise, and blur), we train models that classify ambiguous images by shape a majority of the time, and outperform baselines on out-of-distribution test sets.

33, TITLE:      GramGAN: Deep 3D Texture Synthesis From 2D Exemplars
http://arxiv.org/abs/2006.16112
AUTHORS:        Tiziano Portenier ; Siavash Bigdeli ; Orcun Goksel
HIGHLIGHT:      We present a novel texture synthesis framework, enabling the generation of infinite, high-quality 3D textures given a 2D exemplar image.

34, TITLE:      Confidence Calibration and Predictive Uncertainty Estimation for Deep Medical Image Segmentation
http://arxiv.org/abs/1911.13273
AUTHORS:        Alireza Mehrtash ; William M. Wells III ; Clare M. Tempany ; Purang Abolmaesumi ; Tina Kapur
COMMENTS:       Journal of IEEE Transactions on Medical Imaging
HIGHLIGHT:      In this paper, we study predictive uncertainty estimation in FCNs for medical image segmentation.

35, TITLE:      Noise2Inverse: Self-supervised deep convolutional denoising for linear image reconstruction
http://arxiv.org/abs/2001.11801
AUTHORS:        Allard A. Hendriksen ; Daniel M. Pelt ; K. Joost Batenburg
HIGHLIGHT:      Here, we propose Noise2Inverse, a deep CNN-based denoising method for linear image reconstruction algorithms that does not require any additional clean or noisy data.

36, TITLE:      PraNet: Parallel Reverse Attention Network for Polyp Segmentation
http://arxiv.org/abs/2006.11392
AUTHORS:        Deng-Ping Fan ; Ge-Peng Ji ; Tao Zhou ; Geng Chen ; Huazhu Fu ; Jianbing Shen ; Ling Shao
COMMENTS:       Accepted to MICCAI 2020
HIGHLIGHT:      To address these challenges, we propose a parallel reverse attention network (PraNet) for accurate polyp segmentation in colonoscopy images.

37, TITLE:      Improving Sequence Tagging for Vietnamese Text Using Transformer-based Neural Models
http://arxiv.org/abs/2006.15994
AUTHORS:        Viet Bui The ; Oanh Tran Thi ; Phuong Le-Hong
HIGHLIGHT:      This paper describes our study on using mutilingual BERT embeddings and some new neural models for improving sequence tagging tasks for the Vietnamese language.

38, TITLE:      A Deep Learning-Based Method for Automatic Segmentation of Proximal Femur from Quantitative Computed Tomography Images
http://arxiv.org/abs/2006.05513
AUTHORS:        Chen Zhao ; Joyce H. Keyak ; Jinshan Tang ; Tadashi S. Kaneko ; Sundeep Khosla ; Shreyasee Amin ; Elizabeth J. Atkinson ; Lan-Juan Zhao ; Michael J. Serou ; Chaoyang Zhang ; Hui Shen ; Hong-Wen Deng ; Weihua Zhou
HIGHLIGHT:      We aim to develop a deep-learning-based method for automatic proximal femur segmentation.

39, TITLE:      Fabric Image Representation Encoding Networks for Large-scale 3D Medical Image Analysis
http://arxiv.org/abs/2006.15578
AUTHORS:        Siyu Liu ; Wei Dai ; Craig Engstrom ; Jurgen Fripp ; Peter B. Greer ; Stuart Crozier ; Jason A. Dowling ; Shekhar S. Chandra
COMMENTS:       12 pages, 10 figures
HIGHLIGHT:      In this work, a novel 3D segmentation network, Fabric Image Representation Networks (FIRENet), is proposed to extract and encode generalisable feature representations from multiple medical image datasets in a large-scale manner.

40, TITLE:      Intrinsic Dynamic Shape Prior for Fast, Sequential and Dense Non-Rigid Structure from Motion with Detection of Temporally-Disjoint Rigidity
http://arxiv.org/abs/1909.02468
AUTHORS:        Vladislav Golyanik ; André Jonas ; Didier Stricker ; Christian Theobalt
COMMENTS:       16 pages, 13 figures, 5 tables
HIGHLIGHT:      To fill this gap, we propose a hybrid approach that extracts prior shape knowledge from an input sequence with NRSfM and uses it as a dynamic shape prior for sequential surface recovery in scenarios with recurrence.

41, TITLE:      MANTRA: A Machine Learning reference lightcurve dataset for astronomical transient event recognition
http://arxiv.org/abs/2006.13163
AUTHORS:        Mauricio Neira ; Catalina Gómez ; John F. Suárez-Pérez ; Diego A. Gómez ; Juan Pablo Reyes ; Marcela Hernández Hoyos ; Pablo Arbeláez ; Jaime E. Forero-Romero
COMMENTS:       ApJS accepted, 17 pages, 14 figures
HIGHLIGHT:      We introduce MANTRA, an annotated dataset of 4869 transient and 71207 non-transient object lightcurves built from the Catalina Real Time Transient Survey.

42, TITLE:      Rethinking Positional Encoding in Language Pre-training
http://arxiv.org/abs/2006.15595
AUTHORS:        Guolin Ke ; Di He ; Tie-Yan Liu
HIGHLIGHT:      In this work, we investigate the problems in the previous formulations and propose a new positional encoding method for BERT called Transformer with Untied Positional Encoding (TUPE).

43, TITLE:      Privacy at Facebook Scale
http://arxiv.org/abs/2006.14109
AUTHORS:        Paulo Tanaka ; Sameet Sapra ; Nikolay Laptev
HIGHLIGHT:      Privacy at Facebook Scale

44, TITLE:      DisCont: Self-Supervised Visual Attribute Disentanglement using Context Vectors
http://arxiv.org/abs/2006.05895
AUTHORS:        Sarthak Bhagat ; Vishaal Udandarao ; Shagun Uppal
COMMENTS:       Published at the 37th International Conference on Machine Learning (ICML 2020) Workshop on ML Interpretability for Scientific Discovery
HIGHLIGHT:      In this paper, we propose a self-supervised framework DisCont to disentangle multiple attributes by exploiting the structural inductive biases within images.

45, TITLE:      Progressive Graph Learning for Open-Set Domain Adaptation
http://arxiv.org/abs/2006.12087
AUTHORS:        Yadan Luo ; Zijian Wang ; Zi Huang ; Mahsa Baktashmotlagh
HIGHLIGHT:      More specifically, we introduce an end-to-end Progressive Graph Learning (PGL) framework where a graph neural network with episodic training is integrated to suppress underlying conditional shift and adversarial learning is adopted to close the gap between the source and target distributions.

46, TITLE:      Improving GAN Training with Probability Ratio Clipping and Sample Reweighting
http://arxiv.org/abs/2006.06900
AUTHORS:        Yue Wu ; Pan Zhou ; Andrew Gordon Wilson ; Eric P. Xing ; Zhiting Hu
COMMENTS:       Fixed typos. Code available at: https://github.com/Holmeswww/PPOGAN
HIGHLIGHT:      We propose a new variational GAN training framework which enjoys superior training stability.

47, TITLE:      Bootstrapping Complete The Look at Pinterest
http://arxiv.org/abs/2006.10792
AUTHORS:        Eileen Li ; Eric Kim ; Andrew Zhai ; Josh Beal ; Kunlong Gu
COMMENTS:       9 pages, 12 figures, To be published in KDD '20
HIGHLIGHT:      In this paper, we will describe how we bootstrapped the Complete The Look (CTL) system at Pinterest. We will introduce our outfit dataset of over 1 million outfits and 4 million objects, a subset of which we will make available to the research community, and describe the pipeline used to obtain and refresh this dataset.

48, TITLE:      Fair Generative Modeling via Weak Supervision
http://arxiv.org/abs/1910.12008
AUTHORS:        Kristy Choi ; Aditya Grover ; Trisha Singh ; Rui Shu ; Stefano Ermon
COMMENTS:       First two authors contributed equally
HIGHLIGHT:      We present a weakly supervised algorithm for overcoming dataset bias for deep generative models.

49, TITLE:      Range Conditioned Dilated Convolutions for Scale Invariant 3D Object Detection
http://arxiv.org/abs/2005.09927
AUTHORS:        Alex Bewley ; Pei Sun ; Thomas Mensink ; Dragomir Anguelov ; Cristian Sminchisescu
COMMENTS:       22 pages
HIGHLIGHT:      This paper presents a novel 3D object detection framework that processes LiDAR data directly on a representation of the sensor's native range images.

50, TITLE:      Extension Complexity, MSO Logic, and Treewidth
http://arxiv.org/abs/1507.04907
AUTHORS:        Petr Kolman ; Martin Koutecký ; Hans Raj Tiwary
HIGHLIGHT:      We consider the convex hull $P_{\varphi}(G)$ of all satisfying assignments of a given MSO formula $\varphi$ on a given graph $G$.

51, TITLE:      Building Rule Hierarchies for Efficient Logical Rule Learning from Knowledge Graphs
http://arxiv.org/abs/2006.16171
AUTHORS:        Yulong Gu ; Yu Guan ; Paolo Missier
HIGHLIGHT:      In this work, we address such scalability issues by proposing new methods for pruning unpromising rules using rule hierarchies.

52, TITLE:      Budgeted Training: Rethinking Deep Neural Network Training Under Resource Constraints
http://arxiv.org/abs/1905.04753
AUTHORS:        Mengtian Li ; Ersin Yumer ; Deva Ramanan
COMMENTS:       ICLR 2020. Project page with code is at http://www.cs.cmu.edu/~mengtial/proj/budgetnn/
HIGHLIGHT:      Therefore, we introduce a formal setting for studying training under the non-asymptotic, resource-constrained regime, i.e., budgeted training.

53, TITLE:      Up or Down? Adaptive Rounding for Post-Training Quantization
http://arxiv.org/abs/2004.10568
AUTHORS:        Markus Nagel ; Rana Ali Amjad ; Mart van Baalen ; Christos Louizos ; Tijmen Blankevoort
COMMENTS:       Published as a conference paper at ICML 2020
HIGHLIGHT:      In this paper, we propose AdaRound, a better weight-rounding mechanism for post-training quantization that adapts to the data and the task loss.

54, TITLE:      Adapting to Unseen Environments through Explicit Representation of Context
http://arxiv.org/abs/2002.05640
AUTHORS:        Cem C. Tutum ; Risto Miikkulainen
COMMENTS:       7 pages, 7 figures, ALife2020
HIGHLIGHT:      This paper proposes a principled approach where a context module is coevolved with a skill module.

55, TITLE:      Implicit Generation and Generalization in Energy-Based Models
http://arxiv.org/abs/1903.08689
AUTHORS:        Yilun Du ; Igor Mordatch
HIGHLIGHT:      We present techniques to scale MCMC based EBM training on continuous neural networks, and we show its success on the high-dimensional data domains of ImageNet32x32, ImageNet128x128, CIFAR-10, and robotic hand trajectories, achieving better samples than other likelihood models and nearing the performance of contemporary GAN approaches, while covering all modes of the data.

56, TITLE:      Cognitive and motor compliance in intentional human-robot interaction
http://arxiv.org/abs/1911.01753
AUTHORS:        Hendry Ferreira Chame ; Jun Tani
COMMENTS:       "\c{opyright} 20XX IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works."
HIGHLIGHT:      We propose a variational model inspired by the principles of predictive coding and active inference to study intentionality and cognitive compliance, and an intermittent control concept for motor deliberation and compliance based on torque feed-back.

57, TITLE:      Mutual Context Network for Jointly Estimating Egocentric Gaze and Actions
http://arxiv.org/abs/1901.01874
AUTHORS:        Yifei Huang ; Zhenqiang Li ; Minjie Cai ; Yoichi Sato
HIGHLIGHT:      In this work, we address two coupled tasks of gaze prediction and action recognition in egocentric videos by exploring their mutual context.

58, TITLE:      From Simulation to Real World Maneuver Execution using Deep Reinforcement Learning
http://arxiv.org/abs/2005.07023
AUTHORS:        Alessandro Paolo Capasso ; Giulio Bacchiani ; Alberto Broggi
COMMENTS:       Accepted at Intelligent Vehicle 2020 (IV2020)
HIGHLIGHT:      In this work, we investigate these problems in the autonomous driving field, especially for a maneuver planning module for roundabout insertions.

59, TITLE:      Understanding when spatial transformer networks do not support invariance, and what to do about it
http://arxiv.org/abs/2004.11678
AUTHORS:        Lukas Finnveden ; Ylva Jansson ; Tony Lindeberg
COMMENTS:       To appear in International Conference on Pattern Recognition (ICPR2020). 12 pages, 7 figures
HIGHLIGHT:      We present a simple proof for this and study the practical implications, showing that this inability is coupled with decreased classification accuracy.

60, TITLE:      Supermasks in Superposition
http://arxiv.org/abs/2006.14769
AUTHORS:        Mitchell Wortsman ; Vivek Ramanujan ; Rosanne Liu ; Aniruddha Kembhavi ; Mohammad Rastegari ; Jason Yosinski ; Ali Farhadi
HIGHLIGHT:      We present the Supermasks in Superposition (SupSup) model, capable of sequentially learning thousands of tasks without catastrophic forgetting.

61, TITLE:      Imbalanced Gradients: A New Cause of Overestimated Adversarial Robustness
http://arxiv.org/abs/2006.13726
AUTHORS:        Linxi Jiang ; Xingjun Ma ; Zejia Weng ; James Bailey ; Yu-Gang Jiang
COMMENTS:       17 pages, 7 figues
HIGHLIGHT:      In this paper, we identify a more subtle situation called \emph{Imbalanced Gradients} that can also cause overestimated adversarial robustness.

62, TITLE:      Empower Entity Set Expansion via Language Model Probing
http://arxiv.org/abs/2004.13897
AUTHORS:        Yunyi Zhang ; Jiaming Shen ; Jingbo Shang ; Jiawei Han
COMMENTS:       ACL 2020
HIGHLIGHT:      In this study, we propose a novel iterative set expansion framework that leverages automatically generated class names to address the semantic drift issue.

63, TITLE:      Automated Pavement Crack Segmentation Using U-Net-based Convolutional Neural Network
http://arxiv.org/abs/2001.01912
AUTHORS:        Stephen L. H. Lau ; Edwin K. P. Chong ; Xu Yang ; Xin Wang
COMMENTS:       Accepted for publication in IEEE Access
HIGHLIGHT:      In this paper, we propose a deep learning technique based on a convolutional neural network to perform segmentation tasks on pavement crack images.

64, TITLE:      Attention-Based Neural Networks for Sentiment Attitude Extraction using Distant Supervision
http://arxiv.org/abs/2006.13730
AUTHORS:        Nicolay Rusnachenko ; Natalia Loukachevitch
COMMENTS:       10 pages, 9 figures. The preprint of an article published in the proceedings of the 10th International Conference on Web Intelligence, Mining and Semantics (WIMS 2020). The final authenticated publication is available online at https://doi.org/10.1145/3405962.3405985. arXiv admin note: substantial text overlap with arXiv:2006.11605
HIGHLIGHT:      In this paper, we provide a study on attention-based context encoders in the sentiment attitude extraction task.

65, TITLE:      Decomposing Word Embedding with the Capsule Network
http://arxiv.org/abs/2004.13844
AUTHORS:        Xin Liu ; Qingcai Chen ; Yan Liu ; Joanna Siebert ; Baotian Hu ; Xiangping Wu ; Buzhou Tang
HIGHLIGHT:      In this paper, we discuss a capsule network-based approach, taking advantage of capsule's potential for recognizing highly overlapping features and dealing with segmentation.

66, TITLE:      Automatic Kernel Generation for Volta Tensor Cores
http://arxiv.org/abs/2006.12645
AUTHORS:        Somashekaracharya G. Bhaskaracharya ; Julien Demouth ; Vinod Grover
HIGHLIGHT:      In this paper, we describe a polyhedral approach to generate efficient CUDA kernels for matrix multiplication using inline assembly instructions for programming tensor cores on NVIDIA Volta GPUs.

67, 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 iUNet, which employs novel learnable and invertible up- and downsampling operations, thereby making the use of memory-efficient backpropagation possible.

68, TITLE:      Seeing Through Fog Without Seeing Fog: Deep Multimodal Sensor Fusion in Unseen Adverse Weather
http://arxiv.org/abs/1902.08913
AUTHORS:        Mario Bijelic ; Tobias Gruber ; Fahim Mannan ; Florian Kraus ; Werner Ritter ; Klaus Dietmayer ; Felix Heide
HIGHLIGHT:      Departing from proposal-level fusion, we propose a single-shot model that adaptively fuses features, driven by measurement entropy. To address this challenge we present a novel multimodal dataset acquired in over 10,000km of driving in northern Europe.

69, TITLE:      Re-translation versus Streaming for Simultaneous Translation
http://arxiv.org/abs/2004.03643
AUTHORS:        Naveen Arivazhagan ; Colin Cherry ; Wolfgang Macherey ; George Foster
COMMENTS:       IWSLT 2020
HIGHLIGHT:      We study a related problem in which revisions to the hypothesis beyond strictly appending words are permitted.

70, TITLE:      AFDet: Anchor Free One Stage 3D Object Detection
http://arxiv.org/abs/2006.12671
AUTHORS:        Runzhou Ge ; Zhuangzhuang Ding ; Yihan Hu ; Yu Wang ; Sijia Chen ; Li Huang ; Yuan Li
COMMENTS:       Accepted on May 6th, 2020 by CVPRW 2020, published on June 7th, 2020; Baseline detector for the 1st place solutions of Waymo Open Dataset Challenges 2020
HIGHLIGHT:      Most previous works try to solve it using anchor-based detection methods which come with two drawbacks: post-processing is relatively complex and computationally expensive; tuning anchor parameters is tricky.

71, TITLE:      Multi-level colonoscopy malignant tissue detection with adversarial CAC-UNet
http://arxiv.org/abs/2006.15954
AUTHORS:        Chuang Zhu ; Ke Mei ; Ting Peng ; Yihao Luo ; Jun Liu ; Ying Wang ; Mulan Jin
COMMENTS:       accepted by Neurocomputing; winner of the MICCAI DigestPath 2019 challenge on colonoscopy tissue segmentation and classification task
HIGHLIGHT:      In this paper, we propose a highly efficient multi-level malignant tissue detection through the designed adversarial CAC-UNet.

72, TITLE:      Large Deformation Diffeomorphic Image Registration with Laplacian Pyramid Networks
http://arxiv.org/abs/2006.16148
AUTHORS:        Tony C. W. Mok ; Albert C. S. Chung
COMMENTS:       Paper accepted by MICCAI 2020
HIGHLIGHT:      In this paper, we propose a deep Laplacian Pyramid Image Registration Network, which can solve the image registration optimization problem in a coarse-to-fine fashion within the space of diffeomorphic maps.

73, TITLE:      Compositional theories for embedded languages
http://arxiv.org/abs/2006.10604
AUTHORS:        Davide Trotta ; Margherita Zorzi
COMMENTS:       20 pages
HIGHLIGHT:      The aim of this paper is to present a flexible fragment of such a type theory, together with its categorical semantics in terms of enriched categories, following previous investigations.

74, TITLE:      On the monotone complexity of the shift operator
http://arxiv.org/abs/1905.10747
AUTHORS:        Igor S. Sergeev
COMMENTS:       7 pages (in English); 7 pages (in Russian); ver. 2: abstract extended, and bibliography updated
HIGHLIGHT:      In particular, we obtain an alternative way to prove the known complexity bound $\Theta(n\log n)$ for the monotone shift operator on $n$ boolean inputs.

75, TITLE:      Learning to cooperate: Emergent communication in multi-agent navigation
http://arxiv.org/abs/2004.01097
AUTHORS:        Ivana Kajić ; Eser Aygün ; Doina Precup
COMMENTS:       Accepted to CogSci 2020
HIGHLIGHT:      Emergent communication in artificial agents has been studied to understand language evolution, as well as to develop artificial systems that learn to communicate with humans.

76, TITLE:      Deep learning for smart fish farming: applications, opportunities and challenges
http://arxiv.org/abs/2004.11848
AUTHORS:        Xinting Yang ; Song Zhang ; Jintao Liu ; Qinfeng Gao ; Shuanglin Dong ; Chao Zhou
COMMENTS:       43 pages, 7 figures
HIGHLIGHT:      In brief, our purpose is to provide researchers and practitioners with a better understanding of the current state of the art of DL in aquaculture, which can provide strong support for the implementation of smart fish farming.

77, TITLE:      Non-Autoregressive Neural Text-to-Speech
http://arxiv.org/abs/1905.08459
AUTHORS:        Kainan Peng ; Wei Ping ; Zhao Song ; Kexin Zhao
COMMENTS:       Published at ICML 2020. (v3 changed paper title)
HIGHLIGHT:      In this work, we propose ParaNet, a non-autoregressive seq2seq model that converts text to spectrogram.

78, TITLE:      Automatic Generation of Efficient Sparse Tensor Format Conversion Routines
http://arxiv.org/abs/2001.02609
AUTHORS:        Stephen Chou ; Fredrik Kjolstad ; Saman Amarasinghe
COMMENTS:       Presented at PLDI 2020
HIGHLIGHT:      This paper shows how to generate code that efficiently converts sparse tensors between disparate storage formats (data layouts) such as CSR, DIA, ELL, and many others.

79, TITLE:      Graph-based, Self-Supervised Program Repair from Diagnostic Feedback
http://arxiv.org/abs/2005.10636
AUTHORS:        Michihiro Yasunaga ; Percy Liang
COMMENTS:       ICML 2020. Code & data available at https://github.com/michiyasunaga/DrRepair
HIGHLIGHT:      In this work, we propose novel solutions to these two challenges.

80, TITLE:      POMDP Modelling for Assessing Hierarchies
http://arxiv.org/abs/1908.07031
AUTHORS:        Weipeng Huang ; Guangyuan Piao ; Raul Moreno ; Neil J. Hurley
HIGHLIGHT:      This motivates us to propose a framework for assessing the quality of hierarchical clustering allocations which covers the case of no ground-truth information.