2020.02.21.txt

==========New Papers==========
1, TITLE:       Intermediate problems in modular circuits satisfiability
http://arxiv.org/abs/2002.08626
AUTHORS:        Paweł M. Idziak ; Piotr Kawałek ; Jacek Krzaczkowski
HIGHLIGHT:      In this paper we provide a broad class of examples, lying in this grey area, and show that, under the Exponential Time Hypothesis and Strong Exponential Size Hypothesis (saying that Boolean circuits need exponentially many modular counting gates to produce boolean conjunctions of any arity), satisfiability over these algebras have intermediate complexity between $\Omega(2^{c\log^{h-1} n})$ and $O(2^{c\log^h n})$, where $h$ measures how much a nilpotent algebra fails to be supernilpotent.

2, TITLE:       Balancing Cost and Benefit with Tied-Multi Transformers
http://arxiv.org/abs/2002.08614
AUTHORS:        Raj Dabre ; Raphael Rubino ; Atsushi Fujita
COMMENTS:       Extended version of our previous manuscript available at arXiv:1908.10118
HIGHLIGHT:      We propose and evaluate a novel procedure for training multiple Transformers with tied parameters which compresses multiple models into one enabling the dynamic choice of the number of encoder and decoder layers during decoding.

3, TITLE:       Guiding attention in Sequence-to-sequence models for Dialogue Act prediction
http://arxiv.org/abs/2002.08801
AUTHORS:        Pierre Colombo ; Emile Chapuis ; Matteo Manica ; Emmanuel Vignon ; Giovanna Varni ; Chloe Clavel
HIGHLIGHT:      In this work, we introduce a seq2seq model tailored for DA classification using: a hierarchical encoder, a novel guided attention mechanism and beam search applied to both training and inference.

4, TITLE:       Contextual Lensing of Universal Sentence Representations
http://arxiv.org/abs/2002.08866
AUTHORS:        Jamie Kiros
COMMENTS:       10 pages
HIGHLIGHT:      In this work we propose Contextual Lensing, a methodology for inducing context-oriented universal sentence vectors.

5, TITLE:       FrameAxis: Characterizing Framing Bias and Intensity with Word Embedding
http://arxiv.org/abs/2002.08608
AUTHORS:        Haewoon Kwak ; Jisun An ; Yong-Yeol Ahn
COMMENTS:       8 pages
HIGHLIGHT:      We propose FrameAxis, a method of characterizing the framing of a given text by identifying the most relevant semantic axes ("microframes") defined by antonym word pairs.

6, TITLE:       Even faster algorithms for CSAT over~supernilpotent algebras
http://arxiv.org/abs/2002.08634
AUTHORS:        Piotr Kawałek ; Jacek Krzaczkowski
HIGHLIGHT:      In this paper two algorithms solving circuit satisfiability problem over supernilpotent algebras are presented.

7, TITLE:       Federated pretraining and fine tuning of BERT using clinical notes from multiple silos
http://arxiv.org/abs/2002.08562
AUTHORS:        Dianbo Liu ; Tim Miller
HIGHLIGHT:      In this article, we show that it is possible to both pretrain and fine tune BERT models in a federated manner using clinical texts from different silos without moving the data.

8, TITLE:       MonoLayout: Amodal scene layout from a single image
http://arxiv.org/abs/2002.08394
AUTHORS:        Kaustubh Mani ; Swapnil Daga ; Shubhika Garg ; N. Sai Shankar ; Krishna Murthy Jatavallabhula ; K. Madhava Krishna
COMMENTS:       To be presented at WACV 2020 Video: https://www.youtube.com/watch?v=HcroGyo6yRQ Project page: https://hbutsuak95.github.io/monolayout
HIGHLIGHT:      In this paper, we address the novel, highly challenging problem of estimating the layout of a complex urban driving scenario.

9, TITLE:       JRMOT: A Real-Time 3D Multi-Object Tracker and a New Large-Scale Dataset
http://arxiv.org/abs/2002.08397
AUTHORS:        Abhijeet Shenoi ; Mihir Patel ; JunYoung Gwak ; Patrick Goebel ; Amir Sadeghian ; Hamid Rezatofighi ; Roberto Martin-Martin ; Silvio Savarese
COMMENTS:       9 pages, 2 figures, 2 tables
HIGHLIGHT:      In this work we present JRMOT, a novel 3D MOT system that integrates information from 2D RGB images and 3D point clouds into a real-time performing framework. As part of our work, we release the JRDB dataset, a novel large scale 2D+3D dataset and benchmark annotated with over 2 million boxes and 3500 time consistent 2D+3D trajectories across 54 indoor and outdoor scenes.

10, TITLE:      Do you comply with AI? -- Personalized explanations of learning algorithms and their impact on employees' compliance behavior
http://arxiv.org/abs/2002.08777
AUTHORS:        NIklas Kuhl ; Jodie Lobana ; Christian Meske
COMMENTS:       Fortieth International Conference on Information Systems (ICIS) 2019, Munich, Germany. All Authors contributed equally in shared first authorship
HIGHLIGHT:      Hence, compliance with the recommendations of such artifacts, which can impact employees' task performance significantly, is still subject to research - and personalization of AI explanations seems to be a promising concept in this regard.

11, TITLE:      A Generalizable Knowledge Framework for Semantic Indoor Mapping Based on Markov Logic Networks and Data Driven MCMC
http://arxiv.org/abs/2002.08402
AUTHORS:        Ziyuan Liu ; Georg von Wichert
HIGHLIGHT:      In this paper, we propose a generalizable knowledge framework for data abstraction, i.e. finding compact abstract model for input data using predefined abstract terms.

12, TITLE:      The Problem with Metrics is a Fundamental Problem for AI
http://arxiv.org/abs/2002.08512
AUTHORS:        Rachel Thomas ; David Uminsky
COMMENTS:       Accepted to EDSC (Ethics of Data Science Conference) 2020
HIGHLIGHT:      The Problem with Metrics is a Fundamental Problem for AI

13, TITLE:      Differential Dynamic Programming Neural Optimizer
http://arxiv.org/abs/2002.08809
AUTHORS:        Guan-Horng Liu ; Tianrong Chen ; Evangelos A. Theodorou
HIGHLIGHT:      In this work, we make an attempt along this line by reformulating the training procedure from the trajectory optimization perspective.

14, TITLE:      A Bayes-Optimal View on Adversarial Examples
http://arxiv.org/abs/2002.08859
AUTHORS:        Eitan Richardson ; Yair Weiss
HIGHLIGHT:      In this paper, we argue for examining adversarial examples from the perspective of Bayes-Optimal classification. We construct realistic image datasets for which the Bayes-Optimal classifier can be efficiently computed and derive analytic conditions on the distributions so that the optimal classifier is either robust or vulnerable.

15, TITLE:      Safe Counterfactual Reinforcement Learning
http://arxiv.org/abs/2002.08536
AUTHORS:        Yusuke Narita ; Shota Yasui ; Kohei Yata
HIGHLIGHT:      We develop a method for predicting the performance of reinforcement learning and bandit algorithms, given historical data that may have been generated by a different algorithm.

16, TITLE:      Deep Learning Estimation of Multi-Tissue Constrained Spherical Deconvolution with Limited Single Shell DW-MRI
http://arxiv.org/abs/2002.08820
AUTHORS:        Vishwesh Nath ; Sudhir K. Pathak ; Kurt G. Schilling ; Walt Schneider ; Bennett A. Landman
COMMENTS:       10 pages, 7 figures
HIGHLIGHT:      Herein, we explore the possibility of using deep learning on single shell data (using the b=1000 s/mm2 from the Human Connectome Project (HCP)) to estimate the information content captured by 8th order MT-CSD using the full three shell data (b=1000, 2000, and 3000 s/mm2 from HCP).

17, TITLE:      Deep Fusion of Local and Non-Local Features for Precision Landslide Recognition
http://arxiv.org/abs/2002.08547
AUTHORS:        Qing Zhu ; Lin Chen ; Han Hu ; Binzhi Xu ; Yeting Zhang ; Haifeng Li
HIGHLIGHT:      Aiming to solve this problem, this paper proposes an effective approach to fuse both local and non-local features to surmount the contextual problem.

18, TITLE:      Fine tuning U-Net for ultrasound image segmentation: which layers?
http://arxiv.org/abs/2002.08438
AUTHORS:        Mina Amiri ; Rupert Brooks ; Hassan Rivaz
HIGHLIGHT:      In this study, we investigated the effect of fine-tuning different layers of a U-Net which was trained on segmentation of natural images in breast ultrasound image segmentation.

19, TITLE:      Cross-stained Segmentation from Renal Biopsy Images Using Multi-level Adversarial Learning
http://arxiv.org/abs/2002.08587
AUTHORS:        Ke Mei ; Chuang Zhu ; Lei Jiang ; Jun Liu ; Yuanyuan Qiao
COMMENTS:       Accepted by ICASSP2020
HIGHLIGHT:      In this paper, we design a robust and flexible model for cross-stained segmentation.

20, TITLE:      Improved Approximate Degree Bounds For k-distinctness
http://arxiv.org/abs/2002.08389
AUTHORS:        Nikhil S. Mande ; Justin Thaler ; Shuchen Zhu
HIGHLIGHT:      For any constant k >= 4, we improve the lower bound to Omega(N^{(3/4)-1/(4k)}).

21, TITLE:      Cutting Corners
http://arxiv.org/abs/2002.08730
AUTHORS:        Ville Salo
COMMENTS:       44 pages, 4 figures
HIGHLIGHT:      We define and study a class of subshifts of finite type (SFTs) defined by a family of allowed patterns of the same shape where, for any contents of the shape minus a corner, the number of ways to fill in the corner is the same.

22, TITLE:      Bounding the expected run-time of nonconvex optimization with early stopping
http://arxiv.org/abs/2002.08856
AUTHORS:        Thomas Flynn ; Kwang Min Yu ; Abid Malik ; Nicolas D'Imperio ; Shinjae Yoo
HIGHLIGHT:      We develop the approach in the general setting of a first-order optimization algorithm, with possibly biased update directions subject to a geometric drift condition.

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

24, TITLE:      Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation
http://arxiv.org/abs/2002.08546
AUTHORS:        Jian Liang ; Dapeng Hu ; Jiashi Feng
HIGHLIGHT:      In this work we tackle a novel setting where only a trained source model is available and investigate how we can effectively utilize such a model without source data to solve UDA problems.

25, TITLE:      Learning Object Scale With Click Supervision for Object Detection
http://arxiv.org/abs/2002.08555
AUTHORS:        Liao Zhang ; Yan Yan ; Lin Cheng ; Hanzi Wang
HIGHLIGHT:      To achieve a good trade-off between annotation cost and object detection performance,we propose a simple yet effective method which incorporatesCNN visualization with click supervision to generate the pseudoground-truths (i.e., bounding boxes).

26, TITLE:      KaoKore: A Pre-modern Japanese Art Facial Expression Dataset
http://arxiv.org/abs/2002.08595
AUTHORS:        Yingtao Tian ; Chikahiko Suzuki ; Tarin Clanuwat ; Mikel Bober-Irizar ; Alex Lamb ; Asanobu Kitamoto
HIGHLIGHT:      To bridge this gap, in this work we propose a new dataset KaoKore which consists of faces extracted from pre-modern Japanese artwork.

27, TITLE:      Focus on Semantic Consistency for Cross-domain Crowd Understanding
http://arxiv.org/abs/2002.08623
AUTHORS:        Tao Han ; Junyu Gao ; Yuan Yuan ; Qi Wang
COMMENTS:       Accpeted by ICASSP2020
HIGHLIGHT:      In this paper, we propose a domain adaptation method to eliminate it.

28, TITLE:      Captioning Images Taken by People Who Are Blind
http://arxiv.org/abs/2002.08565
AUTHORS:        Danna Gurari ; Yinan Zhao ; Meng Zhang ; Nilavra Bhattacharya
HIGHLIGHT:      Observing that people who are blind have relied on (human-based) image captioning services to learn about images they take for nearly a decade, we introduce the first image captioning dataset to represent this real use case.

29, TITLE:      Optimizing Black-box Metrics with Adaptive Surrogates
http://arxiv.org/abs/2002.08605
AUTHORS:        Qijia Jiang ; Olaoluwa Adigun ; Harikrishna Narasimhan ; Mahdi Milani Fard ; Maya Gupta
HIGHLIGHT:      We pose the training problem as an optimization over a relaxed surrogate space, which we solve by estimating local gradients for the metric and performing inexact convex projections.

30, TITLE:      On the Versatility of Open Logical Relations: Continuity, Automatic Differentiation, and a Containment Theorem
http://arxiv.org/abs/2002.08489
AUTHORS:        Gilles Barthe ; Raphaëlle Crubillé ; Ugo Dal Lago ; Francesco Gavazzo
HIGHLIGHT:      To overcome this issue, we study a generalization of the concept of a logical relation, called \emph{open logical relation}, and prove that it can be fruitfully applied in several contexts in which the property of interest is about expressions of first-order type.

31, TITLE:      Learning to Walk in the Real World with Minimal Human Effort
http://arxiv.org/abs/2002.08550
AUTHORS:        Sehoon Ha ; Peng Xu ; Zhenyu Tan ; Sergey Levine ; Jie Tan
HIGHLIGHT:      In this paper, we develop a system for learning legged locomotion policies with deep RL in the real world with minimal human effort.

32, TITLE:      Soundness conditions for big-step semantics
http://arxiv.org/abs/2002.08738
AUTHORS:        Francesco Dagnino ; Viviana Bono ; Elena Zucca ; Mariangiola Dezani-Ciancaglini
HIGHLIGHT:      We propose a general proof technique to show that a predicate is sound, that is, prevents stuck computation, with respect to a big-step semantics.

33, TITLE:      Expressing Objects just like Words: Recurrent Visual Embedding for Image-Text Matching
http://arxiv.org/abs/2002.08510
AUTHORS:        Tianlang Chen ; Jiebo Luo
COMMENTS:       Accepted by AAAI-20
HIGHLIGHT:      To address this problem, we propose a Dual Path Recurrent Neural Network (DP-RNN) which processes images and sentences symmetrically by recurrent neural networks (RNN).

34, TITLE:      Cooperative LIDAR Object Detection via Feature Sharing in Deep Networks
http://arxiv.org/abs/2002.08440
AUTHORS:        Ehsan Emad Marvasti ; Arash Raftari ; Amir Emad Marvasti ; Yaser P. Fallah ; Rui Guo ; HongSheng Lu
COMMENTS:       7 pages, 6 figures
HIGHLIGHT:      In this paper, we aim to mitigate the effects of these limitations by introducing the concept of feature sharing for cooperative object detection (FS-COD).

35, TITLE:      SD-GAN: Structural and Denoising GAN reveals facial parts under occlusion
http://arxiv.org/abs/2002.08448
AUTHORS:        Samik Banerjee ; Sukhendu Das
COMMENTS:       Recommended for revision in Neurocomputing, Elsevier
HIGHLIGHT:      In this paper, we propose a generative model to reconstruct the missing parts of the face which are under occlusion.

36, 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 ; Bjoern 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.

37, TITLE:      Table-Top Scene Analysis Using Knowledge-Supervised MCMC
http://arxiv.org/abs/2002.08417
AUTHORS:        Ziyuan Liu ; Dong Chen ; Kai M. Wurm ; Georg von Wichert
HIGHLIGHT:      In this paper, we propose a probabilistic method to generate abstract scene graphs for table-top scenes from 6D object pose estimates.

38, TITLE:      Modelling response to trypophobia trigger using intermediate layers of ImageNet networks
http://arxiv.org/abs/2002.08490
AUTHORS:        Piotr Woźnicki ; Michał Kuźba ; Piotr Migdał
COMMENTS:       3 pages, 2 figures, 1 table
HIGHLIGHT:      In this paper, we approach the problem of detecting trypophobia triggers using Convolutional neural networks.

39, TITLE:      Detecting Code Clones with Graph Neural Networkand Flow-Augmented Abstract Syntax Tree
http://arxiv.org/abs/2002.08653
AUTHORS:        Wenhan Wang ; Ge Li ; Bo Ma ; Xin Xia ; Zhi Jin
COMMENTS:       Accepted by SANER 2020
HIGHLIGHT:      To leverage control and data flow information, in this paper, we build a graph representation of programs called flow-augmented abstract syntax tree (FA-AST).

40, TITLE:      Interactive Natural Language-based Person Search
http://arxiv.org/abs/2002.08434
AUTHORS:        Vikram Shree ; Wei-Lun Chao ; Mark Campbell
COMMENTS:       8 pages, 12 figures, Published in IEEE Robotics and Automation Letters (RA-L), "Dataset at: https://github.com/vikshree/QA_PersonSearchLanguageData" , Video attachment at: https://www.youtube.com/watch?v=Yyxu8uVUREE&feature=youtu.be
HIGHLIGHT:      In this work, we consider the problem of searching people in an unconstrained environment, with natural language descriptions.

41, TITLE:      An empirical study of Conv-TasNet
http://arxiv.org/abs/2002.08688
AUTHORS:        Berkan Kadioglu ; Michael Horgan ; Xiaoyu Liu ; Jordi Pons ; Dan Darcy ; Vivek Kumar
COMMENTS:       In proceedings of ICASSP2020
HIGHLIGHT:      In this paper, we conduct an empirical study of Conv-TasNet and propose an enhancement to the encoder/decoder that is based on a (deep) non-linear variant of it.

42, TITLE:      Disentangled Speech Embeddings using Cross-modal Self-supervision
http://arxiv.org/abs/2002.08742
AUTHORS:        Arsha Nagrani ; Joon Son Chung ; Samuel Albanie ; Andrew Zisserman
COMMENTS:       To appear in ICASSP 2020. The first three authors contributed equally to this work
HIGHLIGHT:      The objective of this paper is to learn representations of speaker identity without access to manually annotated data.

43, TITLE:      Imputer: Sequence Modelling via Imputation and Dynamic Programming
http://arxiv.org/abs/2002.08926
AUTHORS:        William Chan ; Chitwan Saharia ; Geoffrey Hinton ; Mohammad Norouzi ; Navdeep Jaitly
COMMENTS:       preprint
HIGHLIGHT:      This paper presents the Imputer, a neural sequence model that generates output sequences iteratively via imputations.

44, TITLE:      Wavesplit: End-to-End Speech Separation by Speaker Clustering
http://arxiv.org/abs/2002.08933
AUTHORS:        Neil Zeghidour ; David Grangier
HIGHLIGHT:      We introduce Wavesplit, an end-to-end speech separation system.

45, TITLE:      Quantum Time-Space Tradeoffs by Recording Queries
http://arxiv.org/abs/2002.08944
AUTHORS:        Yassine Hamoudi ; Frédéric Magniez
COMMENTS:       17 pages
HIGHLIGHT:      We use the recording queries technique of Zhandry [Zha19] to prove lower bounds in the exponentially small success probability regime, with applications to time-space tradeoffs.

46, TITLE:      Automatic Gesture Recognition in Robot-assisted Surgery with Reinforcement Learning and Tree Search
http://arxiv.org/abs/2002.08718
AUTHORS:        Xiaojie Gao ; Yueming Jin ; Qi Dou ; Pheng-Ann Heng
COMMENTS:       Accepted as a conference paper in ICRA 2020
HIGHLIGHT:      In this paper, we propose a framework based on reinforcement learning and tree search for joint surgical gesture segmentation and classification.

47, TITLE:      A Comprehensive Scoping Review of Bayesian Networks in Healthcare: Past, Present and Future
http://arxiv.org/abs/2002.08627
AUTHORS:        Evangelia Kyrimi ; Scott McLachlan ; Kudakwashe Dube ; Mariana R. Neves ; Ali Fahmi ; Norman Fenton
HIGHLIGHT:      To map the way forward, the paper proposes future research directions and makes recommendations regarding BN development methods and adoption in practice.

48, TITLE:      Measuring Social Biases in Grounded Vision and Language Embeddings
http://arxiv.org/abs/2002.08911
AUTHORS:        Candace Ross ; Boris Katz ; Andrei Barbu
HIGHLIGHT:      We introduce the space of generalizations (Grounded-WEAT and Grounded-SEAT) and demonstrate that three generalizations answer different yet important questions about how biases, language, and vision interact.

49, TITLE:      The Fluidity of Concept Representations in Human Brain Signals
http://arxiv.org/abs/2002.08880
AUTHORS:        Eva Hendrikx ; Lisa Beinborn
COMMENTS:       12 pages, 5 figures, 1 table
HIGHLIGHT:      In this work, we analyze the discriminability of concrete and abstract concepts in fMRI data using a range of analysis methods.

50, TITLE:      MEUZZ: Smart Seed Scheduling for Hybrid Fuzzing
http://arxiv.org/abs/2002.08568
AUTHORS:        Yaohui Chen ; Mansour Ahmadi ; Reza Mirzazade farkhani ; Boyu Wang ; Long Lu
HIGHLIGHT:      To overcome this problem, we design a Machine learning-Enhanced hybrid fUZZing system (MEUZZ), which employs supervised machine learning for adaptive and generalizable seed scheduling.

51, TITLE:      Unsupervised Multi-Class Domain Adaptation: Theory, Algorithms, and Practice
http://arxiv.org/abs/2002.08681
AUTHORS:        Yabin Zhang ; Bin Deng ; Hui Tang ; Lei Zhang ; Kui Jia
COMMENTS:       The journal manuscript extended significantly from our preliminary CVPR conference paper. Codes are available at: https://github.com/YBZh/MultiClassDA
HIGHLIGHT:      In this paper, we study the formalism of unsupervised multi-class domain adaptation (multi-class UDA), which underlies some recent algorithms whose learning objectives are only motivated empirically.

52, TITLE:      Unsupervised Domain Adaptation via Discriminative Manifold Embedding and Alignment
http://arxiv.org/abs/2002.08675
AUTHORS:        You-Wei Luo ; Chuan-Xian Ren ; Pengfei Ge ; Ke-Kun Huang ; Yu-Feng Yu
COMMENTS:       Accepted to AAAI 2020
HIGHLIGHT:      In this paper, a Riemannian manifold learning framework is proposed to achieve transferability and discriminability consistently.

53, TITLE:      sKPNSGA-II: Knee point based MOEA with self-adaptive angle for Mission Planning Problems
http://arxiv.org/abs/2002.08867
AUTHORS:        Cristian Ramirez-Atencia ; Sanaz Mostaghim ; David Camacho
COMMENTS:       Submitted to Applied Soft Computing
HIGHLIGHT:      This paper presents a new algorithm that has been designed to obtain the most significant solutions from the Pareto Optimal Frontier (POF).

54, TITLE:      Learn to Design the Heuristics for Vehicle Routing Problem
http://arxiv.org/abs/2002.08539
AUTHORS:        Lei Gao ; Mingxiang Chen ; Qichang Chen ; Ganzhong Luo ; Nuoyi Zhu ; Zhixin Liu
COMMENTS:       10 pages, 6 figures
HIGHLIGHT:      This paper presents an approach to learn the local-search heuristics that iteratively improves the solution of Vehicle Routing Problem (VRP).

55, TITLE:      The (Generalized) Orthogonality Dimension of (Generalized) Kneser Graphs: Bounds and Applications
http://arxiv.org/abs/2002.08580
AUTHORS:        Alexander Golovnev ; Ishay Haviv
COMMENTS:       19 pages
HIGHLIGHT:      The contribution of the present work is two-folded.

56, TITLE:      Algorithms and Lower Bounds for de Morgan Formulas of Low-Communication Leaf Gates
http://arxiv.org/abs/2002.08533
AUTHORS:        Valentine Kabanets ; Sajin Koroth ; Zhenjian Lu ; Dimitrios Myrisiotis ; Igor Oliveira
HIGHLIGHT:      We give lower bounds and (SAT, Learning, and PRG) algorithms for $FORMULA[n^{1.99}]\circ \mathcal{G}$, for classes $\mathcal{G}$ of functions with low communication complexity.

57, TITLE:      Stochastic Regret Minimization in Extensive-Form Games
http://arxiv.org/abs/2002.08493
AUTHORS:        Gabriele Farina ; Christian Kroer ; Tuomas Sandholm
HIGHLIGHT:      In this paper we develop a new framework for developing stochastic regret minimization methods.

58, TITLE:      Uncovering Coresets for Classification With Multi-Objective Evolutionary Algorithms
http://arxiv.org/abs/2002.08645
AUTHORS:        Pietro Barbiero ; Giovanni Squillero ; Alberto Tonda
COMMENTS:       9 pages, 3 figures, conference. Submitted to ICML 2020
HIGHLIGHT:      Building on previous works, a novel approach is presented: candidate corsets are iteratively optimized, adding and removing samples.

59, TITLE:      A Novel Framework for Selection of GANs for an Application
http://arxiv.org/abs/2002.08641
AUTHORS:        Tanya Motwani ; Manojkumar Parmar
COMMENTS:       23 pages, 1 figures, 7 tables
HIGHLIGHT:      We propose a novel framework to identify candidate GANs for a specific use case based on architecture, loss, regularization and divergence.

60, TITLE:      Boosting Adversarial Training with Hypersphere Embedding
http://arxiv.org/abs/2002.08619
AUTHORS:        Tianyu Pang ; Xiao Yang ; Yinpeng Dong ; Kun Xu ; Hang Su ; Jun Zhu
HIGHLIGHT:      In order to promote the reliability of the adversarially trained models, we propose to boost AT via incorporating hypersphere embedding (HE), which can regularize the adversarial features onto compact hypersphere manifolds.

61, TITLE:      Constructive Game Logic
http://arxiv.org/abs/2002.08523
AUTHORS:        Brandon Bohrer ; André Platzer
COMMENTS:       74 pages, extended preprint for ESOP
HIGHLIGHT:      Our major contributions include: 1) a novel realizability semantics capturing the adversarial dynamics of games, 2) a natural deduction calculus and operational semantics describing the computational meaning of strategies via proof-terms, and 3) theoretical results including soundness of the proof calculus w.r.t. realizability semantics, progress and preservation of the operational semantics of proofs, and Existence Properties on support of the extraction of computational artifacts from game proofs.

62, TITLE:      Decomposing Probabilistic Lambda-calculi
http://arxiv.org/abs/2002.08392
AUTHORS:        Ugo Dal Lago ; Giulio Guerrieri ; Willem Heijltjes
HIGHLIGHT:      We present a probabilistic lambda-calculus where the probabilistic operator is decomposed into two syntactic constructs: a generator, which represents a probabilistic event; and a consumer, which acts on the term depending on a given event.

63, TITLE:      Multi-Agent Reinforcement Learning as a Computational Tool for Language Evolution Research: Historical Context and Future Challenges
http://arxiv.org/abs/2002.08878
AUTHORS:        Clément Moulin-Frier ; Pierre-Yves Oudeyer
HIGHLIGHT:      The goal of this paper is to position recent MARL contributions within the historical context of language evolution research, as well as to extract from this theoretical and computational background a few challenges for future research.

64, TITLE:      Spatiotemporal Relationship Reasoning for Pedestrian Intent Prediction
http://arxiv.org/abs/2002.08945
AUTHORS:        Bingbin Liu ; Ehsan Adeli ; Zhangjie Cao ; Kuan-Hui Lee ; Abhijeet Shenoi ; Adrien Gaidon ; Juan Carlos Niebles
COMMENTS:       Accepted at ICRA 2020 and IEEE Robotics and Automation Letters
HIGHLIGHT:      In this paper, we present a framework based on graph convolution to uncover the spatiotemporal relationships in the scene for reasoning about pedestrian intent. In addition, we introduce a new dataset designed specifically for autonomous-driving scenarios in areas with dense pedestrian populations: the Stanford-TRI Intent Prediction (STIP) dataset.

65, TITLE:      Deep Learning-Based Feature Extraction in Iris Recognition: Use Existing Models, Fine-tune or Train From Scratch?
http://arxiv.org/abs/2002.08916
AUTHORS:        Aidan Boyd ; Adam Czajka ; Kevin Bowyer
COMMENTS:       Presented at BTAS 2019
HIGHLIGHT:      In this work we explore five different sets of weights for the popular ResNet-50 architecture to find out whether iris-specific feature extractors perform better than models trained for non-iris tasks.

66, 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 removing shortcut features automatically.

67, TITLE:      Strategy to Increase the Safety of a DNN-based Perception for HAD Systems
http://arxiv.org/abs/2002.08935
AUTHORS:        Timo Sämann ; Peter Schlicht ; Fabian Hüger
HIGHLIGHT:      The aim of this paper is to present a framework for the description and mitigation of DNN insufficiencies and the derivation of relevant safety mechanisms to increase the safety of DNNs.

68, TITLE:      Bimodal Distribution Removal and Genetic Algorithm in Neural Network for Breast Cancer Diagnosis
http://arxiv.org/abs/2002.08729
AUTHORS:        Ke Quan
HIGHLIGHT:      Multiple linear programming models have been devised to approximate the relationship between cell features and tumour malignancy.

69, TITLE:      AdvMS: A Multi-source Multi-cost Defense Against Adversarial Attacks
http://arxiv.org/abs/2002.08439
AUTHORS:        Xiao Wang ; Siyue Wang ; Pin-Yu Chen ; Xue Lin ; Peter Chin
COMMENTS:       Accepted by 45th International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2020)
HIGHLIGHT:      In this paper, we study principles of designing multi-source and multi-cost schemes where defense performance is boosted from multiple defending components.

70, TITLE:      Contextual Equivalence for Signal Flow Graphs
http://arxiv.org/abs/2002.08874
AUTHORS:        Filippo Bonchi ; Robin Piedeleu ; Pawel Sobocinski ; Fabio Zanasi
COMMENTS:       Accepted for publication in the proceedings of the 23rd International Conference on Foundations of Software Science and Computation Structures (FoSSaCS 2020)
HIGHLIGHT:      We extend the signal flow calculus---a compositional account of the classical signal flow graph model of computation---to encompass affine behaviour, and furnish it with a novel operational semantics.

71, TITLE:      I-SPEC: An End-to-End Framework for Learning Transportable, Shift-Stable Models
http://arxiv.org/abs/2002.08948
AUTHORS:        Adarsh Subbaswamy ; Suchi Saria
HIGHLIGHT:      In this paper, we propose I-SPEC, an end-to-end framework that addresses this shortcoming by using data to learn a partial ancestral graph (PAG).

72, TITLE:      T-Net: A Template-Supervised Network for Task-specific Feature Extraction in Biomedical Image Analysis
http://arxiv.org/abs/2002.08406
AUTHORS:        Weinan Song ; Yuan Liang ; Kun Wang ; Lei He
HIGHLIGHT:      In this paper, we propose a template-supervised network T-Net for task-specific feature extraction.

73, TITLE:      Bi-directional Dermoscopic Feature Learning and Multi-scale Consistent Decision Fusion for Skin Lesion Segmentation
http://arxiv.org/abs/2002.08694
AUTHORS:        Xiaohong Wang ; Xudong Jiang ; Henghui Ding ; Jun Liu
COMMENTS:       Accepted to TIP
HIGHLIGHT:      In this paper, we propose a novel bi-directional dermoscopic feature learning (biDFL) framework to model the complex correlation between skin lesions and their informative context.

74, TITLE:      Roto-Translation Equivariant Convolutional Networks: Application to Histopathology Image Analysis
http://arxiv.org/abs/2002.08725
AUTHORS:        Maxime W. Lafarge ; Erik J. Bekkers ; Josien P. W. Pluim ; Remco Duits ; Mitko Veta
HIGHLIGHT:      We propose a framework to encode the geometric structure of the special Euclidean motion group SE(2) in convolutional networks to yield translation and rotation equivariance via the introduction of SE(2)-group convolution layers.

75, TITLE:      Stroke Constrained Attention Network for Online Handwritten Mathematical Expression Recognition
http://arxiv.org/abs/2002.08670
AUTHORS:        Jiaming Wang ; Jun Du ; Jianshu Zhang
HIGHLIGHT:      In this paper, we propose a novel stroke constrained attention network (SCAN) which treats stroke as the basic unit for encoder-decoder based online handwritten mathematical expression recognition (HMER).

76, TITLE:      Photorealistic Lip Sync with Adversarial Temporal Convolutional Networks
http://arxiv.org/abs/2002.08700
AUTHORS:        Ruobing Zheng ; Zhou Zhu ; Bo Song ; Changjiang Ji
COMMENTS:       9 pages, 7 figures
HIGHLIGHT:      In this paper, we present a novel lip-sync solution for producing a high-quality and photorealistic talking head from speech.

77, TITLE:      Neural Network Compression Framework for fast model inference
http://arxiv.org/abs/2002.08679
AUTHORS:        Alexander Kozlov ; Ivan Lazarevich ; Vasily Shamporov ; Nikolay Lyalyushkin ; Yury Gorbachev
COMMENTS:       9 pages, 1 figure
HIGHLIGHT:      In this work we present a new framework for neural networks compression with fine-tuning, which we called Neural Network Compression Framework (NNCF).

78, TITLE:      A survey on Semi-, Self- and Unsupervised Techniques in Image Classification
http://arxiv.org/abs/2002.08721
AUTHORS:        Lars Schmarje ; Monty Santarossa ; Simon-Martin Schröder ; Reinhard Koch
COMMENTS:       Submitted to IJCV
HIGHLIGHT:      In this survey we provide an overview of often used techniques and methods in image classification with fewer labels.

79, TITLE:      Object 6D Pose Estimation with Non-local Attention
http://arxiv.org/abs/2002.08749
AUTHORS:        Jianhan Mei ; Henghui Ding ; Xudong Jiang
HIGHLIGHT:      In this paper, we address the challenging task of estimating 6D object pose from a single RGB image.

80, TITLE:      Scalable Constrained Bayesian Optimization
http://arxiv.org/abs/2002.08526
AUTHORS:        David Eriksson ; Matthias Poloczek
HIGHLIGHT:      We propose the scalable constrained Bayesian optimization (SCBO) algorithm that addresses the above challenges by data-independent transformations of the functions and follows the recent theme of local Bayesian optimization.

81, TITLE:      On the Uniqueness of Simultaneous Rational Function Reconstruction
http://arxiv.org/abs/2002.08748
AUTHORS:        Eleonora Guerrini ; Romain Lebreton ; Ilaria Zappatore
HIGHLIGHT:      In this work, we prove that uniqueness is guaranteed for a generic instance.



==========Updates to Previous Papers==========
1, TITLE:       Language GANs Falling Short
http://arxiv.org/abs/1811.02549
AUTHORS:        Massimo Caccia ; Lucas Caccia ; William Fedus ; Hugo Larochelle ; Joelle Pineau ; Laurent Charlin
HIGHLIGHT:      In this work, we make several surprising observations which contradict common beliefs.

2, TITLE:       Multilogue-Net: A Context Aware RNN for Multi-modal Emotion Detection and Sentiment Analysis in Conversation
http://arxiv.org/abs/2002.08267
AUTHORS:        Aman Shenoy ; Ashish Sardana
COMMENTS:       10 pages, 4 figures, 6 tables
HIGHLIGHT:      In this paper, we propose a recurrent neural network architecture that attempts to take into account all the mentioned drawbacks, and keeps track of the context of the conversation, interlocutor states, and the emotions conveyed by the speakers in the conversation.

3, TITLE:       Bridging the Gap Between Computational Photography and Visual Recognition
http://arxiv.org/abs/1901.09482
AUTHORS:        Rosaura G. VidalMata ; Sreya Banerjee ; Brandon RichardWebster ; Michael Albright ; Pedro Davalos ; Scott McCloskey ; Ben Miller ; Asong Tambo ; Sushobhan Ghosh ; Sudarshan Nagesh ; Ye Yuan ; Yueyu Hu ; Junru Wu ; Wenhan Yang ; Xiaoshuai Zhang ; Jiaying Liu ; Zhangyang Wang ; Hwann-Tzong Chen ; Tzu-Wei Huang ; Wen-Chi Chin ; Yi-Chun Li ; Mahmoud Lababidi ; Charles Otto ; Walter J. Scheirer
COMMENTS:       CVPR Prize Challenge: http://www.ug2challenge.org
HIGHLIGHT:      We introduce six new algorithms for image restoration or enhancement, which were created as part of the IARPA sponsored UG^2 Challenge workshop held at CVPR 2018. To address this, we introduce the UG^2 dataset as a large-scale benchmark composed of video imagery captured under challenging conditions, and two enhancement tasks designed to test algorithmic impact on visual quality and automatic object recognition. Furthermore, we propose a set of metrics to evaluate the joint improvement of such tasks as well as individual algorithmic advances, including a novel psychophysics-based evaluation regime for human assessment and a realistic set of quantitative measures for object recognition performance.

4, TITLE:       Working Memory Graphs
http://arxiv.org/abs/1911.07141
AUTHORS:        Ricky Loynd ; Roland Fernandez ; Asli Celikyilmaz ; Adith Swaminathan ; Matthew Hausknecht
COMMENTS:       8 pages, 7 figures, 8 page appendix
HIGHLIGHT:      We present the Working Memory Graph (WMG), an agent that employs multi-head self-attention to reason over a dynamic set of vectors representing observed and recurrent state.

5, TITLE:       Information-geometric optimization with natural selection
http://arxiv.org/abs/1912.03395
AUTHORS:        Jakub Otwinowski ; Colin LaMont
COMMENTS:       changed title
HIGHLIGHT:      Evolutionary algorithms, inspired by natural evolution, aim to optimize difficult objective functions without computing derivatives.

6, TITLE:       Perception-oriented Single Image Super-Resolution via Dual Relativistic Average Generative Adversarial Networks
http://arxiv.org/abs/1911.03464
AUTHORS:        Yuan Ma ; Kewen Liu ; Hongxia Xiong ; Panpan Fang ; Xiaojun Li ; Yalei Chen ; Chaoyang Liu
COMMENTS:       Re-submit after codes reviewing
HIGHLIGHT:      Regarding the issues, this paper develops a perception-oriented single image SR algorithm via dual relativistic average generative adversarial networks.

7, TITLE:       Single Image Super-resolution via Dense Blended Attention Generative Adversarial Network for Clinical Diagnosis
http://arxiv.org/abs/1906.06575
AUTHORS:        Kewen Liu ; Yuan Ma ; Hongxia Xiong ; Zejun Yan ; Zhijun Zhou ; Chaoyang Liu ; Panpan Fang ; Xiaojun Li ; Yalei Chen
COMMENTS:       We abandoned this paper due to its limitation only applied on medical images, please view our lastest work at arXiv:1911.03464
HIGHLIGHT:      In order to address the issue that medical images would suffer from severe blurring caused by lack of high-frequency details, this paper develops a novel image super-resolution(SR) algorithm called SR-DBAN via dense neural network and blended attention mechanism.

8, TITLE:       Harmonization of diffusion MRI datasets with adaptive dictionary learning
http://arxiv.org/abs/1910.00272
AUTHORS:        Samuel St-Jean ; Max A. Viergever ; Alexander Leemans
COMMENTS:       v4: Peer review round 2 v3: Peer reviewed version v2: Fix minor text issue + add supp materials v1: To be submitted to Neuroimage
HIGHLIGHT:      In this work, we propose a new harmonization algorithm based on adaptive dictionary learning to mitigate the unwanted variability caused by different scanner hardware while preserving the natural biological variability present in the data.

9, TITLE:       ViP: Virtual Pooling for Accelerating CNN-based Image Classification and Object Detection
http://arxiv.org/abs/1906.07912
AUTHORS:        Zhuo Chen ; Jiyuan Zhang ; Ruizhou Ding ; Diana Marculescu
COMMENTS:       8 pages
HIGHLIGHT:      In this paper, we propose Virtual Pooling (ViP), a model-level approach to improve speed and energy consumption of CNN-based image classification and object detection tasks, with a provable error bound.

10, TITLE:      Efficient Solvers for Sparse Subspace Clustering
http://arxiv.org/abs/1804.06291
AUTHORS:        Farhad Pourkamali-Anaraki ; James Folberth ; Stephen Becker
COMMENTS:       This paper is accepted for publication in Signal Processing
HIGHLIGHT:      For both $\ell_1$ and $\ell_0$, algorithms to compute the proximity operator in the presence of affine constraints have not been presented in the SSC literature, so we derive an exact and efficient algorithm that solves the $\ell_1$ case with just $O(n^2)$ flops.

11, TITLE:      Improving 3D U-Net for Brain Tumor Segmentation by Utilizing Lesion Prior
http://arxiv.org/abs/1907.00281
AUTHORS:        Po-Yu Kao ; Jefferson W. Chen ; B. S. Manjunath
COMMENTS:       5 pages, 4 figures, 1 table, LNCS format
HIGHLIGHT:      We propose a novel, simple and effective method to integrate lesion prior and a 3D U-Net for improving brain tumor segmentation.

12, TITLE:      Revisiting Self-Training for Neural Sequence Generation
http://arxiv.org/abs/1909.13788
AUTHORS:        Junxian He ; Jiatao Gu ; Jiajun Shen ; Marc'Aurelio Ranzato
COMMENTS:       ICLR 2020. The first two authors contributed equally
HIGHLIGHT:      To further encourage this mechanism, we propose to inject noise to the input space, resulting in a "noisy" version of self-training.

13, TITLE:      Learning First-Order Symbolic Representations for Planning from the Structure of the State Space
http://arxiv.org/abs/1909.05546
AUTHORS:        Blai Bonet ; Hector Geffner
COMMENTS:       Proc. ECAI-2020
HIGHLIGHT:      In this work we address this split by showing how the first-order symbolic representations that are used by planners can be learned from non-symbolic inputs that encode the structure of the state space.

14, TITLE:      Towards a complete 3D morphable model of the human head
http://arxiv.org/abs/1911.08008
AUTHORS:        Stylianos Ploumpis ; Evangelos Ververas ; Eimear O' Sullivan ; Stylianos Moschoglou ; Haoyang Wang ; Nick Pears ; William A. P. Smith ; Baris Gecer ; Stefanos Zafeiriou
COMMENTS:       18 pages, 18 figures, submitted to Transactions on Pattern Analysis and Machine Intelligence (TPAMI) on the 9th of October as an extension paper of the original oral CVPR paper : arXiv:1903.03785
HIGHLIGHT:      To achieve this, we propose two methods for combining existing 3DMMs of different overlapping head parts: i. use a regressor to complete missing parts of one model using the other, ii.

15, TITLE:      Sperm Detection and Tracking in Phase-Contrast Microscopy Image Sequences using Deep Learning and Modified CSR-DCF
http://arxiv.org/abs/2002.04034
AUTHORS:        Mohammad reza Mohammadi ; Mohammad Rahimzadeh ; Abolfazl Attar
HIGHLIGHT:      In this article, we used a deep fully convolutional network, as the object detector.

16, TITLE:      Fast Efficient Object Detection Using Selective Attention
http://arxiv.org/abs/1811.07502
AUTHORS:        Shivanthan Yohanandan ; Andy Song ; Adrian G. Dyer ; Angela Faragasso ; Subhrajit Roy ; Dacheng Tao
COMMENTS:       Retraction due to significant oversight
HIGHLIGHT:      Fast Efficient Object Detection Using Selective Attention

17, TITLE:      Adversarial Filters of Dataset Biases
http://arxiv.org/abs/2002.04108
AUTHORS:        Ronan Le Bras ; Swabha Swayamdipta ; Chandra Bhagavatula ; Rowan Zellers ; Matthew E. Peters ; Ashish Sabharwal ; Yejin Choi
HIGHLIGHT:      We investigate one recently proposed approach, AFLite, which adversarially filters such dataset biases, as a means to mitigate the prevalent overestimation of machine performance.

18, TITLE:      Stabilizing DARTS with Amended Gradient Estimation on Architectural Parameters
http://arxiv.org/abs/1910.11831
AUTHORS:        Kaifeng Bi ; Changping Hu ; Lingxi Xie ; Xin Chen ; Longhui Wei ; Qi Tian
COMMENTS:       21 pages, 11 figures, submitted to ICML 2020, extensive results are added
HIGHLIGHT:      Our approach bridges the gap from two aspects, namely, amending the estimation on the architectural gradients, and unifying the hyper-parameter settings in the search and re-training stages.

19, TITLE:      Identifying and Compensating for Feature Deviation in Imbalanced Deep Learning
http://arxiv.org/abs/2001.01385
AUTHORS:        Han-Jia Ye ; Hong-You Chen ; De-Chuan Zhan ; Wei-Lun Chao
HIGHLIGHT:      We propose to incorporate class-dependent temperatures (CDT) in learning a ConvNet: CDT forces the minor-class instances to have larger decision values in training, so as to compensate for the effect of feature deviation in testing.

20, TITLE:      Visually Guided Self Supervised Learning of Speech Representations
http://arxiv.org/abs/2001.04316
AUTHORS:        Abhinav Shukla ; Konstantinos Vougioukas ; Pingchuan Ma ; Stavros Petridis ; Maja Pantic
COMMENTS:       Accepted at ICASSP 2020 v2: Updated to the ICASSP 2020 camera ready version
HIGHLIGHT:      We propose a framework for learning audio representations guided by the visual modality in the context of audiovisual speech.

21, TITLE:      Robustness of accelerated first-order algorithms for strongly convex optimization problems
http://arxiv.org/abs/1905.11011
AUTHORS:        Hesameddin Mohammadi ; Meisam Razaviyayn ; Mihailo R. Jovanović
COMMENTS:       45 pages, 6 figures
HIGHLIGHT:      Specifically, for unconstrained, smooth, strongly convex optimization problems, we examine the mean-squared error in the optimization variable when the iterates are perturbed by additive white noise.

22, TITLE:      Comprehensive Taxonomies of Nature- and Bio-inspired Optimization: Inspiration versus Algorithmic Behavior, Critical Analysis and Recommendations
http://arxiv.org/abs/2002.08136
AUTHORS:        Daniel Molina ; Javier Poyatos ; Javier Del Ser ; Salvador García ; Amir Hussain ; Francisco Herrera
COMMENTS:       76 pages, 6 figures
HIGHLIGHT:      In recent years, a great variety of nature- and bio-inspired algorithms has been reported in the literature.

23, TITLE:      A comparison of Vector Symbolic Architectures
http://arxiv.org/abs/2001.11797
AUTHORS:        Kenny Schlegel ; Peer Neubert ; Peter Protzel
COMMENTS:       9 pages, 3 figures, preprint - manuscript
HIGHLIGHT:      In this paper, we give an overview of eight available VSA implementations and discuss their commonalities and differences in the underlying vector space, bundling, and binding/unbinding operations.

24, TITLE:      The Limitations of Stylometry for Detecting Machine-Generated Fake News
http://arxiv.org/abs/1908.09805
AUTHORS:        Tal Schuster ; Roei Schuster ; Darsh J Shah ; Regina Barzilay
COMMENTS:       Accepted for Computational Linguistics journal (squib). Previously posted with title "Are We Safe Yet? The Limitations of Distributional Features for Fake News Detection"
HIGHLIGHT:      In light of these concerns, several studies have proposed to detect machine-generated fake news by capturing their stylistic differences from human-written text. We create two benchmarks demonstrating the stylistic similarity between malicious and legitimate uses of LMs, employed in auto-completion and editing-assistance settings.

25, TITLE:      Tree Search vs Optimization Approaches for Map Generation
http://arxiv.org/abs/1903.11678
AUTHORS:        Debosmita Bhaumik ; Ahmed Khalifa ; Michael Cerny Green ; Julian Togelius
COMMENTS:       10 pages, 9 figures, submitted to FDG 2020
HIGHLIGHT:      For purposes of comparison, we use a simplified map generation problem where only passable and impassable tiles exist, three different map representations, and a set of objectives that are representative of those commonly found in actual level generation problem.

26, TITLE:      Mining Uncertain Event Data in Process Mining
http://arxiv.org/abs/1910.00089
AUTHORS:        Marco Pegoraro ; Wil M. P. van der Aalst
COMMENTS:       18 pages, 7 figures, 3 tables
HIGHLIGHT:      In this paper we analyze the previously unexplored setting of uncertain event logs: logs where quantified uncertainty is recorded together with the corresponding data.

27, TITLE:      Precise neural network computation with imprecise analog devices
http://arxiv.org/abs/1606.07786
AUTHORS:        Jonathan Binas ; Daniel Neil ; Giacomo Indiveri ; Shih-Chii Liu ; Michael Pfeiffer
HIGHLIGHT:      We propose a framework that exploits the power of deep learning to compensate for this mismatch by incorporating the measured device variations as constraints in the neural network training process.

28, TITLE:      Coresets for the Nearest-Neighbor Rule
http://arxiv.org/abs/2002.06650
AUTHORS:        Alejandro Flores-Velazco ; David M. Mount
HIGHLIGHT:      In this paper, we address these shortcomings by proposing new approximation-sensitive criteria for the nearest-neighbor condensation problem, along with practical algorithms with provable performance guarantees.

29, TITLE:      Mixup Inference: Better Exploiting Mixup to Defend Adversarial Attacks
http://arxiv.org/abs/1909.11515
AUTHORS:        Tianyu Pang ; Kun Xu ; Jun Zhu
COMMENTS:       ICLR 2020
HIGHLIGHT:      Inspired by simple geometric intuition, we develop an inference principle, named mixup inference (MI), for mixup-trained models.

30, TITLE:      Node Masking: Making Graph Neural Networks Generalize and Scale Better
http://arxiv.org/abs/2001.07524
AUTHORS:        Pushkar Mishra ; Aleksandra Piktus ; Gerard Goossen ; Fabrizio Silvestri
HIGHLIGHT:      In this paper, we discuss some theoretical tools to better visualize the operations performed by state of the art spatial GNNs.

31, TITLE:      A Calculus for Modular Loop Acceleration
http://arxiv.org/abs/2001.01516
AUTHORS:        Florian Frohn
HIGHLIGHT:      In contrast, we present a calculus that allows for combining acceleration techniques in a modular way and we show how to integrate many existing acceleration techniques into our calculus.

32, TITLE:      Classification and Disease Localization in Histopathology Using Only Global Labels: A Weakly-Supervised Approach
http://arxiv.org/abs/1802.02212
AUTHORS:        Pierre Courtiol ; Eric W. Tramel ; Marc Sanselme ; Gilles Wainrib
HIGHLIGHT:      We propose a method for disease localization in the context of weakly supervised learning, where only image-level labels are available during training.

33, TITLE:      HMS-Net: Hierarchical Multi-scale Sparsity-invariant Network for Sparse Depth Completion
http://arxiv.org/abs/1808.08685
AUTHORS:        Zixuan Huang ; Junming Fan ; Shenggan Cheng ; Shuai Yi ; Xiaogang Wang ; Hongsheng Li
COMMENTS:       IEEE Trans. on Image Processing
HIGHLIGHT:      Our extensive experiments and component analysis on two public benchmarks, KITTI depth completion benchmark and NYU-depth-v2 dataset, demonstrate the effectiveness of the proposed approach.

34, TITLE:      ExpandNets: Linear Over-parameterization to Train Compact Convolutional Networks
http://arxiv.org/abs/1811.10495
AUTHORS:        Shuxuan Guo ; Jose M. Alvarez ; Mathieu Salzmann
HIGHLIGHT:      In this paper, we introduce an approach to training a given compact network.

35, TITLE:      Founded (Auto)Epistemic Equilibrium Logic Satisfies Epistemic Splitting
http://arxiv.org/abs/1907.09247
AUTHORS:        Jorge Fandinno
COMMENTS:       Paper presented at the 35th International Conference on Logic Programming (ICLP 2019), Las Cruces, New Mexico, USA, 20-25 September 2019, 16 pages
HIGHLIGHT:      In this paper, we prove that FAEEL also satisfies the epistemic splitting property something that, together with foundedness, was not fulfilled by any other approach up to date.

36, TITLE:      Multimodal feature fusion for CNN-based gait recognition: an empirical comparison
http://arxiv.org/abs/1806.07753
AUTHORS:        Francisco Manuel Castro ; Manuel Jesús Marín-Jiménez ; Nicolás Guil ; Nicolás Pérez de la Blanca
COMMENTS:       arXiv admin note: text overlap with arXiv:1603.01006
HIGHLIGHT:      In contrast, in this paper we focus on the raw pixels, or simple functions derived from them, letting advanced learning techniques to extract relevant features.

37, TITLE:      FastV2C-HandNet: Fast Voxel to Coordinate Hand Pose Estimation with 3D Convolutional Neural Networks
http://arxiv.org/abs/1907.06327
AUTHORS:        Rohan Lekhwani ; Bhupendra Singh
COMMENTS:       13 pages, 5 figures, 2 tables
HIGHLIGHT:      In this paper, we present a novel approach to estimate 3D hand joint locations from 2D depth images.

38, TITLE:      Towards Photographic Image Manipulation with Balanced Growing of Generative Autoencoders
http://arxiv.org/abs/1904.06145
AUTHORS:        Ari Heljakka ; Arno Solin ; Juho Kannala
COMMENTS:       WACV 2020
HIGHLIGHT:      We present a generative autoencoder that provides fast encoding, faithful reconstructions (eg.

39, TITLE:      Deep Network Classification by Scattering and Homotopy Dictionary Learning
http://arxiv.org/abs/1910.03561
AUTHORS:        John Zarka ; Louis Thiry ; Tomás Angles ; Stéphane Mallat
HIGHLIGHT:      We introduce a sparse scattering deep convolutional neural network, which provides a simple model to analyze properties of deep representation learning for classification.

40, TITLE:      Uncertainty-guided Continual Learning with Bayesian Neural Networks
http://arxiv.org/abs/1906.02425
AUTHORS:        Sayna Ebrahimi ; Mohamed Elhoseiny ; Trevor Darrell ; Marcus Rohrbach
COMMENTS:       Accepted at ICLR 2020
HIGHLIGHT:      In contrast, we propose Uncertainty-guided Continual Bayesian Neural Networks (UCB), where the learning rate adapts according to the uncertainty defined in the probability distribution of the weights in networks.

41, TITLE:      Kleene stars of the plane, polylogarithms and symmetries
http://arxiv.org/abs/1811.09091
AUTHORS:        Gérard Henry Edmond Duchamp ; Vincel Hoang Ngoc Minh ; Ngo Quoc Hoan
HIGHLIGHT:      We extend the definition and construct several bases for polylogarithms Li T , where T are some series, recognizable by a finite state (multiplicity) automaton of alphabet 4 X = {x 0 , x 1 }.