2020.06.04.txt

==========New Papers==========
1, TITLE:       NewtonianVAE: Proportional Control and Goal Identification from Pixels via Physical Latent Spaces
http://arxiv.org/abs/2006.01959
AUTHORS:        Miguel Jaques ; Michael Burke ; Timothy Hospedales
HIGHLIGHT:      We introduce a latent dynamics learning framework that is uniquely designed to induce proportional controlability in the latent space, thus enabling the use of much simpler controllers than prior work.

2, TITLE:       Ear2Face: Deep Biometric Modality Mapping
http://arxiv.org/abs/2006.01943
AUTHORS:        Dogucan Yaman ; Fevziye Irem Eyiokur ; Hazım Kemal Ekenel
COMMENTS:       13 pages, 4 figures
HIGHLIGHT:      In this paper, we explore the correlation between different visual biometric modalities.

3, TITLE:       Continual Learning of Predictive Models in Video Sequences via Variational Autoencoders
http://arxiv.org/abs/2006.01945
AUTHORS:        Damian Campo ; Giulia Slavic ; Mohamad Baydoun ; Lucio Marcenaro ; Carlo Regazzoni
COMMENTS:       Manuscript accepted at the 27th IEEE International Conference on Image Processing (ICIP 2020)
HIGHLIGHT:      This paper proposes a method for performing continual learning of predictive models that facilitate the inference of future frames in video sequences.

4, TITLE:       Nurse is Closer to Woman than Surgeon? Mitigating Gender-Biased Proximities in Word Embeddings
http://arxiv.org/abs/2006.01938
AUTHORS:        Vaibhav Kumar ; Tenzin Singhay Bhotia ; Vaibhav Kumar ; Tanmoy Chakraborty
COMMENTS:       TACL 2020
HIGHLIGHT:      In this paper, we propose RAN-Debias, a novel gender debiasing methodology which not only eliminates the bias present in a word vector but also alters the spatial distribution of its neighbouring vectors, achieving a bias-free setting while maintaining minimal semantic offset.

5, TITLE:       Offline and Online Satisfaction Prediction in Open-Domain Conversational Systems
http://arxiv.org/abs/2006.01921
AUTHORS:        Jason Ingyu Choi ; Ali Ahmadvand ; Eugene Agichtein
COMMENTS:       Published in CIKM '19, 10 pages
HIGHLIGHT:      To accomplish this goal, we propose a conversational satisfaction prediction model specifically designed for open-domain spoken conversational agents, called ConvSAT.

6, TITLE:       The Convolution Exponential and Generalized Sylvester Flows
http://arxiv.org/abs/2006.01910
AUTHORS:        Emiel Hoogeboom ; Victor Garcia Satorras ; Jakub M. Tomczak ; Max Welling
HIGHLIGHT:      This paper introduces a new method to build linear flows, by taking the exponential of a linear transformation.

7, TITLE:       Quantifying the Effects of Prosody Modulation on User Engagement and Satisfaction in Conversational Systems
http://arxiv.org/abs/2006.01916
AUTHORS:        Jason Ingyu Choi ; Eugene Agichtein
COMMENTS:       Published in CHIIR 2020, 4 pages
HIGHLIGHT:      In this paper, we investigate one natural approach to this problem, of modulating response prosody, i.e., changing the pitch and cadence of the response to indicate delight, sadness or other common emotions, as well as using pre-recorded interjections.

8, TITLE:       On the Predictive Power of Neural Language Models for Human Real-Time Comprehension Behavior
http://arxiv.org/abs/2006.01912
AUTHORS:        Ethan Gotlieb Wilcox ; Jon Gauthier ; Jennifer Hu ; Peng Qian ; Roger Levy
COMMENTS:       To Appear at CogSci 2020
HIGHLIGHT:      We find that across model architectures and training dataset sizes the relationship between word log-probability and reading time is (near-)linear.

9, TITLE:       AI-Powered Learning: Making Education Accessible, Affordable, and Achievable
http://arxiv.org/abs/2006.01908
AUTHORS:        Ashok Goel
COMMENTS:       17 pages
HIGHLIGHT:      We have developed an AI-powered socio-technical system for making online learning in higher education more accessible, affordable and achievable.

10, TITLE:      On a Class of Constrained Synchronization Problems in NP
http://arxiv.org/abs/2006.01903
AUTHORS:        Stefan Hoffmann
COMMENTS:       arXiv admin note: text overlap with arXiv:2005.05907
HIGHLIGHT:      In this work, we take a closer look at them.

11, TITLE:      Automatic Text Summarization of COVID-19 Medical Research Articles using BERT and GPT-2
http://arxiv.org/abs/2006.01997
AUTHORS:        Virapat Kieuvongngam ; Bowen Tan ; Yiming Niu
HIGHLIGHT:      Our model provides abstractive and comprehensive information based on keywords extracted from the original articles.

12, TITLE:      PILArNet: Public Dataset for Particle Imaging Liquid Argon Detectors in High Energy Physics
http://arxiv.org/abs/2006.01993
AUTHORS:        Corey Adams ; Kazuhiro Terao ; Taritree Wongjirad
HIGHLIGHT:      In this paper we describe the dataset, tasks, and the method used to procure the sample.

13, TITLE:      Training End-to-End Analog Neural Networks with Equilibrium Propagation
http://arxiv.org/abs/2006.01981
AUTHORS:        Jack Kendall ; Ross Pantone ; Kalpana Manickavasagam ; Yoshua Bengio ; Benjamin Scellier
HIGHLIGHT:      We introduce a principled method to train end-to-end analog neural networks by stochastic gradient descent.

14, TITLE:      Quantifying the Uncertainty in Model Parameters Using Gaussian Process-Based Markov Chain Monte Carlo: An Application to Cardiac Electrophysiological Models
http://arxiv.org/abs/2006.01983
AUTHORS:        Jwala Dhamala ; John L. Sapp ; B. Milan Horácek ; Linwei Wang
HIGHLIGHT:      In this paper, we address this issue by integrating surrogate modeling into Metropolis Hasting (MH) sampling of the exact posterior pdfs to improve its acceptance rate.

15, TITLE:      From two rolling shutters to one global shutter
http://arxiv.org/abs/2006.01964
AUTHORS:        Cenek Albl ; Zuzana Kukelova ; Viktor Larsson ; Tomas Pajdla ; Konrad Schindler
COMMENTS:       CVPR 2020
HIGHLIGHT:      We derive equations that describe the underlying geometry for general and special motions and present an efficient method for finding their solutions.

16, TITLE:      The Typology of Polysemy: A Multilingual Distributional Framework
http://arxiv.org/abs/2006.01966
AUTHORS:        Ella Rabinovich ; Yang Xu ; Suzanne Stevenson
COMMENTS:       CogSci 2020 (Annual Meeting of the Cognitive Science Society)
HIGHLIGHT:      We present a novel computational framework that quantifies semantic affinity, the cross-linguistic similarity of lexical semantics for a concept.

17, TITLE:      Characterizing an Analogical Concept Memory for Newellian Cognitive Architectures
http://arxiv.org/abs/2006.01962
AUTHORS:        Shiwali Mohan ; Matt Klenk ; Matthew Shreve ; Kent Evans ; Aaron Ang ; John Maxwell
COMMENTS:       Under review at the Eighth Annual Conference on Advances in Cognitive Systems (ACS 2020)
HIGHLIGHT:      We propose a new long-term declarative memory for Soar that leverages the computational models of analogical reasoning and generalization.

18, TITLE:      Grafted network for person re-identification
http://arxiv.org/abs/2006.01967
AUTHORS:        Jiabao Wang ; Yang Li ; Yang Li ; Zhuang Miao ; Rui Zhang
HIGHLIGHT:      In order to relieve this problem, we propose a novel grafted network (GraftedNet), which is designed by grafting a high-accuracy rootstock and a light-weighted scion.

19, TITLE:      REL: An Entity Linker Standing on the Shoulders of Giants
http://arxiv.org/abs/2006.01969
AUTHORS:        Johannes M. van Hulst ; Faegheh Hasibi ; Koen Dercksen ; Krisztian Balog ; Arjen P. de Vries
HIGHLIGHT:      The REL system presented in this paper aims to fill that gap.

20, TITLE:      On tensor rank and commuting matrices
http://arxiv.org/abs/2006.02374
AUTHORS:        Pascal Koiran
HIGHLIGHT:      In this paper we propose a generalization of the approach used by Strassen in the proof of his $3n/2$ border rank lower bound.

21, TITLE:      Towards Large-Scale Data Mining for Data-Driven Analysis of Sign Languages
http://arxiv.org/abs/2006.02120
AUTHORS:        Boris Mocialov ; Graham Turner ; Helen Hastie
COMMENTS:       https://colab.research.google.com/drive/118Sx1ua-NXy9kjqWi94vz-RrRVYKmnDl?usp=sharing
HIGHLIGHT:      Using our data collection pipeline, we collect and examine the interpretation of songs in both the American Sign Language (ASL) and the Brazilian Sign Language (Libras).

22, TITLE:      Optimizing Neural Networks via Koopman Operator Theory
http://arxiv.org/abs/2006.02361
AUTHORS:        Akshunna S. Dogra ; William T Redman
HIGHLIGHT:      In this work, we take the first steps in making use of this connection.

23, TITLE:      From Probability to Consilience: How Explanatory Values Implement Bayesian Reasoning
http://arxiv.org/abs/2006.02359
AUTHORS:        Zachary Wojtowicz ; Simon DeDeo
COMMENTS:       19 pages, 1 figure, comments welcome
HIGHLIGHT:      We propose a Bayesian account of how these values fit together to guide explanation.

24, TITLE:      From Real to Synthetic and Back: Synthesizing Training Data for Multi-Person Scene Understanding
http://arxiv.org/abs/2006.02110
AUTHORS:        Igor Kviatkovsky ; Nadav Bhonker ; Gerard Medioni
HIGHLIGHT:      We present a method for synthesizing naturally looking images of multiple people interacting in a specific scenario.

25, TITLE:      CNN-based Speed Detection Algorithm for Walking and Running using Wrist-worn Wearable Sensors
http://arxiv.org/abs/2006.02348
AUTHORS:        Venkata Devesh Reddy Seethi ; Pratool Bharti
COMMENTS:       6 pages, 7 figures
HIGHLIGHT:      In this paper, we design, implement and evaluate a convolutional neural network based algorithm that leverages accelerometer and gyroscope sensory data from the wrist-worn device to detect the speed with high precision.

26, TITLE:      Self-Supervised Localisation between Range Sensors and Overhead Imagery
http://arxiv.org/abs/2006.02108
AUTHORS:        Tim Y. Tang ; Daniele De Martini ; Shangzhe Wu ; Paul Newman
COMMENTS:       Accepted to Robotics: Science and Systems (RSS) 2020
HIGHLIGHT:      We present a learned metric localisation method that not only handles the modality difference, but is cheap to train, learning in a self-supervised fashion without metrically accurate ground truth.

27, TITLE:      Exploiting Class Labels to Boost Performance on Embedding-based Text Classification
http://arxiv.org/abs/2006.02104
AUTHORS:        Arkaitz Zubiaga
HIGHLIGHT:      To make the most of these embeddings as features and to boost the performance of classifiers using them, we introduce a weighting scheme, Term Frequency-Category Ratio (TF-CR), which can weight high-frequency, category-exclusive words higher when computing word embeddings.

28, TITLE:      Automatic Setting of DNN Hyper-Parameters by Mixing Bayesian Optimization and Tuning Rules
http://arxiv.org/abs/2006.02105
AUTHORS:        Michele Fraccaroli ; Evelina Lamma ; Fabrizio Riguzzi
HIGHLIGHT:      For this goal, we build a new algorithm for evaluating and analyzing the results of the network on the training and validation sets and use a set of tuning rules to add new hyper-parameters and/or to reduce the hyper-parameter search space to select a better combination.

29, TITLE:      Non-Euclidean Universal Approximation
http://arxiv.org/abs/2006.02341
AUTHORS:        Anastasis Kratsios ; Ievgen Bilokopytov
COMMENTS:       21 Pages
HIGHLIGHT:      We present general conditions describing feature and readout maps that preserve an architecture's ability to approximate any continuous functions uniformly on compacts.

30, TITLE:      Flexible Bayesian Modelling for Nonlinear Image Registration
http://arxiv.org/abs/2006.02338
AUTHORS:        Mikael Brudfors ; Yaël Balbastre ; Guillaume Flandin ; Parashkev Nachev ; John Ashburner
COMMENTS:       Accepted for MICCAI 2020
HIGHLIGHT:      We describe a diffeomorphic registration algorithm that allows groups of images to be accurately aligned to a common space, which we intend to incorporate into the SPM software.

31, TITLE:      Scene relighting with illumination estimation in the latent space on an encoder-decoder scheme
http://arxiv.org/abs/2006.02333
AUTHORS:        Alexandre Pierre Dherse ; Martin Nicolas Everaert ; Jakub Jan Gwizdała
COMMENTS:       Report for the CS-413 project at EPFL, Switzerland
HIGHLIGHT:      In this report we present methods that we tried to achieve that goal.

32, TITLE:      DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
http://arxiv.org/abs/2006.02334
AUTHORS:        Siyuan Qiao ; Liang-Chieh Chen ; Alan Yuille
HIGHLIGHT:      In this paper, we explore this mechanism in the backbone design for object detection.

33, TITLE:      Learning Multi-Modal Nonlinear Embeddings: Performance Bounds and an Algorithm
http://arxiv.org/abs/2006.02330
AUTHORS:        Semih Kaya ; Elif Vural
HIGHLIGHT:      Learning Multi-Modal Nonlinear Embeddings: Performance Bounds and an Algorithm

34, TITLE:      Efficient refinements on YOLOv3 for real-time detection and assessment of diabetic foot Wagner grades
http://arxiv.org/abs/2006.02322
AUTHORS:        Aifu Han ; Yongze Zhang ; Ajuan Li ; Changjin Li ; Fengying Zhao ; Qiujie Dong ; Qin Liu ; Yanting Liu ; Ximei Shen ; Sunjie Yan ; Shengzong Zhou
COMMENTS:       14 pages with 11 figures
HIGHLIGHT:      In this study, we proposed the real-time detection and location method for Wagner grades of DF based on refinements on YOLOv3.

35, TITLE:      A quest for a fair schedule: The Young Physicists' Tournament
http://arxiv.org/abs/2006.02184
AUTHORS:        Katarína Cechlárová ; Ágnes Cseh ; Zsuzsanna Jankó ; Marián Kireš ; Lukáš Miňo
HIGHLIGHT:      Besides formalizing these feasibility conditions, in this paper we formulate several additional fairness conditions for tournament schedules.

36, TITLE:      Multi-Temporal Scene Classification and Scene Change Detection with Correlation based Fusion
http://arxiv.org/abs/2006.02176
AUTHORS:        Lixiang Ru ; Bo Du ; Chen Wu
COMMENTS:       submitted
HIGHLIGHT:      In this work, we proposed a CorrFusion module that fuses the highly correlated components in bi-temporal feature embeddings.

37, TITLE:      CompGuessWhat?!: A Multi-task Evaluation Framework for Grounded Language Learning
http://arxiv.org/abs/2006.02174
AUTHORS:        Alessandro Suglia ; Ioannis Konstas ; Andrea Vanzo ; Emanuele Bastianelli ; Desmond Elliott ; Stella Frank ; Oliver Lemon
COMMENTS:       Accepted to the Annual Conference of the Association for Computational Linguistics (ACL) 2020
HIGHLIGHT:      To remedy this, we present GROLLA, an evaluation framework for Grounded Language Learning with Attributes with three sub-tasks: 1) Goal-oriented evaluation; 2) Object attribute prediction evaluation; and 3) Zero-shot evaluation.

38, TITLE:      A Mixed Initiative Semantic Web Framework for Process Composition
http://arxiv.org/abs/2006.02168
AUTHORS:        Jinghai Rao ; Dimitar Dimitrov ; Paul Hofmann ; Norman Sadeh
HIGHLIGHT:      In this article, we argue that this assumption is often unrealistic.

39, TITLE:      Multi-Agent Cross-Translated Diversification for Unsupervised Machine Translation
http://arxiv.org/abs/2006.02163
AUTHORS:        Xuan-Phi Nguyen ; Shafiq Joty ; Wu Kui ; Ai Ti Aw
HIGHLIGHT:      This work introduces another component to this framework: Multi-Agent Cross-translated Diversification (MACD).

40, TITLE:      Interpolation-based semi-supervised learning for object detection
http://arxiv.org/abs/2006.02158
AUTHORS:        Jisoo Jeong ; Vikas Verma ; Minsung Hyun ; Juho Kannala ; Nojun Kwak
HIGHLIGHT:      In this paper, we propose an Interpolation-based Semi-supervised learning method for object Detection (ISD), which considers and solves the problems caused by applying conventional Interpolation Regularization (IR) directly to object detection.

41, TITLE:      Transfer Learning for British Sign Language Modelling
http://arxiv.org/abs/2006.02144
AUTHORS:        Boris Mocialov ; Graham Turner ; Helen Hastie
COMMENTS:       10 pages, 3 figures
HIGHLIGHT:      In this paper, we examine two transfer learning techniques of fine-tuning and layer substitution for language modelling of British Sign Language.

42, TITLE:      Self-supervised Training of Graph Convolutional Networks
http://arxiv.org/abs/2006.02380
AUTHORS:        Qikui Zhu ; Bo Du ; Pingkun Yan
HIGHLIGHT:      To further improve the learning capacity and model performance under the limited training data, in this paper, we propose two types of self-supervised learning strategies to exploit available information from the input graph structure data itself.

43, TITLE:      CNN Denoisers As Non-Local Filters: The Neural Tangent Denoiser
http://arxiv.org/abs/2006.02379
AUTHORS:        Julián Tachella ; Junqi Tang ; Mike Davies
HIGHLIGHT:      We introduce a novel interpretation of denoising networks with no clean training data in the context of the neural tangent kernel (NTK), elucidating the strong links with well-known non-local filtering techniques, such as non-local means or BM3D.

44, TITLE:      Tangles: a new paradigm for clusters and types
http://arxiv.org/abs/2006.01830
AUTHORS:        Reinhard Diestel
COMMENTS:       arXiv admin note: substantial text overlap with arXiv:1907.07341
HIGHLIGHT:      Tangles: a new paradigm for clusters and types

45, TITLE:      Automatic Differentiation for All Photons Imaging to See Inside Volumetric Scattering Media
http://arxiv.org/abs/2006.01897
AUTHORS:        Tomohiro Maeda ; Ankit Ranjan ; Ramesh Raskar
HIGHLIGHT:      We propose a new framework using automatic differentiation for All Photons Imaging through homogeneous scattering media with unknown optical properties for non-invasive sensing and diagnostics.

46, TITLE:      Learning to Branch for Multi-Task Learning
http://arxiv.org/abs/2006.01895
AUTHORS:        Pengsheng Guo ; Chen-Yu Lee ; Daniel Ulbricht
COMMENTS:       Accepted at ICML 2020
HIGHLIGHT:      In this work, we present an automated multi-task learning algorithm that learns where to share or branch within a network, designing an effective network topology that is directly optimized for multiple objectives across tasks.

47, TITLE:      Adversarial Item Promotion: Vulnerabilities at the Core of Top-N Recommenders that Use Images to Address Cold Start
http://arxiv.org/abs/2006.01888
AUTHORS:        Zhuoran Liu ; Martha larson
COMMENTS:       Our code is available at https://github.com/liuzrcc/AIP
HIGHLIGHT:      In this paper, we demonstrate how unscrupulous merchants can create item images that artificially promote their products, improving their rankings.

48, TITLE:      Generating Random Logic Programs Using Constraint Programming
http://arxiv.org/abs/2006.01889
AUTHORS:        Paulius Dilkas ; Vaishak Belle
HIGHLIGHT:      We present a novel approach to generating random logic programs and random probabilistic logic programs using constraint programming, introducing a new constraint to control the independence structure of the underlying probability distribution.

49, TITLE:      Event Arguments Extraction via Dilate Gated Convolutional Neural Network with Enhanced Local Features
http://arxiv.org/abs/2006.01854
AUTHORS:        Zhigang Kan ; Linbo Qiao ; Sen Yang ; Feng Liu ; Feng Huang
HIGHLIGHT:      In this work, we proposed a novel Event Extraction approach based on multi-layer Dilate Gated Convolutional Neural Network (EE-DGCNN) which has fewer parameters.

50, TITLE:      Aligning Superhuman AI and Human Behavior: Chess as a Model System
http://arxiv.org/abs/2006.01855
AUTHORS:        Reid McIlroy-Young ; Siddhartha Sen ; Jon Kleinberg ; Ashton Anderson
HIGHLIGHT:      We develop and introduce Maia, a customized version of Alpha-Zero trained on human chess games, that predicts human moves at a much higher accuracy than existing engines, and can achieve maximum accuracy when predicting decisions made by players at a specific skill level in a tuneable way.

51, TITLE:      Kernel Taylor-Based Value Function Approximation for Continuous-State Markov Decision Processes
http://arxiv.org/abs/2006.02008
AUTHORS:        Junhong Xu ; Kai Yin ; Lantao Liu
COMMENTS:       Accepted by Robotics: Science and Systems 2020
HIGHLIGHT:      We propose a principled kernel-based policy iteration algorithm to solve the continuous-state Markov Decision Processes (MDPs).

52, TITLE:      Open-Set Recognition with Gaussian Mixture Variational Autoencoders
http://arxiv.org/abs/2006.02003
AUTHORS:        Alexander Cao ; Yuan Luo ; Diego Klabjan
COMMENTS:       12 pages including 8 figures and 4 tables, plus 6 pages of supplementary material
HIGHLIGHT:      In contrast, we train our model to cooperatively learn reconstruction and perform class-based clustering in the latent space.

53, TITLE:      Proximity-based Networking: Small world overlays optimized with particle swarm optimization
http://arxiv.org/abs/2006.02006
AUTHORS:        Chase Smith ; Alex Rusnak
HIGHLIGHT:      We propose a networking scheme that incorporates geographic location in chord for the organization of peers within each node's partitioned key space.

54, TITLE:      MultiNet: Multiclass Multistage Multimodal Motion Prediction
http://arxiv.org/abs/2006.02000
AUTHORS:        Nemanja Djuric ; Henggang Cui ; Zhaoen Su ; Shangxuan Wu ; Huahua Wang ; Fang-Chieh Chou ; Luisa San Martin ; Song Feng ; Rui Hu ; Yang Xu ; Alyssa Dayan ; Sidney Zhang ; Brian C. Becker ; Gregory P. Meyer ; Carlos Vallespi-Gonzalez ; Carl K. Wellington
HIGHLIGHT:      To address this task we propose MultiNet, an end-to-end approach for detection and motion prediction based directly on lidar sensor data.

55, TITLE:      Controlling the Size of Supercompiled Programs using Multi-result Supercompilation
http://arxiv.org/abs/2006.02204
AUTHORS:        Dimitur Krustev
COMMENTS:       Extended version of paper submitted to VPT 2020
HIGHLIGHT:      We consider an approach for controlling result size, based on a combination of multi-result supercompilation and a specific generalization strategy, which avoids code duplication.

56, TITLE:      DGSAC: Density Guided Sampling and Consensus
http://arxiv.org/abs/2006.02413
AUTHORS:        Lokender Tiwari ; Saket Anand
COMMENTS:       Working article
HIGHLIGHT:      In this paper, we deviate from the mode seeking and time budget framework.

57, TITLE:      Emergent Multi-Agent Communication in the Deep Learning Era
http://arxiv.org/abs/2006.02419
AUTHORS:        Angeliki Lazaridou ; Marco Baroni
HIGHLIGHT:      We review language emergence studies from each of these two angles in turn.

58, TITLE:      GFPNet: A Deep Network for Learning Shape Completion in Generic Fitted Primitives
http://arxiv.org/abs/2006.02098
AUTHORS:        Tiberiu Cocias ; Alexandru Razvant ; Sorin Grigorescu
COMMENTS:       8 pages, 14 figures, IEEE Robotics and Automation Letters. Preprint Version. Accepted May, 2020
HIGHLIGHT:      In this paper, we propose an object reconstruction apparatus that uses the so-called Generic Primitives (GP) to complete shapes.

59, TITLE:      Constraint Reductions
http://arxiv.org/abs/2006.02081
AUTHORS:        Olivier Bailleux ; Yacine Boufkhad
HIGHLIGHT:      After recalling its context, we outline a classification of Constraints with respect to their deductive power regarding General Arc Consistency (GAC).

60, TITLE:      An ExpTime Upper Bound for $\mathcal{ALC}$ with Integers (Extended Version)
http://arxiv.org/abs/2006.02078
AUTHORS:        Nadia Labai ; Magdalena Ortiz ; Mantas Šimkus
COMMENTS:       This is a pre-print containing the proofs omitted from the conference version. 36 pages, 3 figures
HIGHLIGHT:      In this paper, we study an extension of $\mathcal{ALC}$ with a rich integer domain that allows for comparisons (between features, and between features and constants coded in unary), and prove that consistency can be solved using automata-theoretic techniques in single exponential time, and thus has no higher worst-case complexity than standard $\mathcal{ALC}$.

61, TITLE:      PLG-IN: Pluggable Geometric Consistency Loss with Wasserstein Distance in Monocular Depth Estimation
http://arxiv.org/abs/2006.02068
AUTHORS:        Noriaki Hirose ; Satoshi Koide ; Keisuke Kawano ; Ruho Kondo
COMMENTS:       9 pages, 6 figures, 2 tables
HIGHLIGHT:      We propose a novel objective to penalize geometric inconsistencies, to improve the performance of depth estimation from monocular camera images.

62, TITLE:      Reference Guided Face Component Editing
http://arxiv.org/abs/2006.02051
AUTHORS:        Qiyao Deng ; Jie Cao ; Yunfan Liu ; Zhenhua Chai ; Qi Li ; Zhenan Sun
HIGHLIGHT:      To break the limitations (e.g. shape, mask or sketch) of the existing methods, we propose a novel framework termed r-FACE (Reference Guided FAce Component Editing) for diverse and controllable face component editing with geometric changes.

63, TITLE:      FBNetV3: Joint Architecture-Recipe Search using Neural Acquisition Function
http://arxiv.org/abs/2006.02049
AUTHORS:        Xiaoliang Dai ; Alvin Wan ; Peizhao Zhang ; Bichen Wu ; Zijian He ; Zhen Wei ; Kan Chen ; Yuandong Tian ; Matthew Yu ; Peter Vajda ; Joseph E. Gonzalez
HIGHLIGHT:      To address this oversight, we present JointNAS to search both (a) architectures and (b) their corresponding training recipes.

64, TITLE:      Fairness-Aware Explainable Recommendation over Knowledge Graphs
http://arxiv.org/abs/2006.02046
AUTHORS:        Zuohui Fu ; Yikun Xian ; Ruoyuan Gao ; Jieyu Zhao ; Qiaoying Huang ; Yingqiang Ge ; Shuyuan Xu ; Shijie Geng ; Chirag Shah ; Yongfeng Zhang ; Gerard de Melo
HIGHLIGHT:      In this paper, we analyze different groups of users according to their level of activity, and find that bias exists in recommendation performance between different groups.

65, TITLE:      Nested Scale Editing for Conditional Image Synthesis
http://arxiv.org/abs/2006.02038
AUTHORS:        Lingzhi Zhang ; Jiancong Wang ; Yinshuang Xu ; Jie Min ; Tarmily Wen ; James C. Gee ; Jianbo Shi
HIGHLIGHT:      We propose an image synthesis approach that provides stratified navigation in the latent code space.

66, TITLE:      Perceiving Unknown in Dark from Perspective of Cell Vibration
http://arxiv.org/abs/2006.02271
AUTHORS:        Xiaozhou Lei ; Minrui Fei ; Wenju Zhou ; Huiyu Zhou
COMMENTS:       13 pages, 17 figures
HIGHLIGHT:      In particular, we here propose a simple yet effective cell vibration energy (CVE) mapping method for image enhancement.

67, TITLE:      FastONN -- Python based open-source GPU implementation for Operational Neural Networks
http://arxiv.org/abs/2006.02267
AUTHORS:        Junaid Malik ; Serkan Kiranyaz ; Moncef Gabbouj
HIGHLIGHT:      This work introduces a fast GPU-enabled library for training operational neural networks, FastONN, which is based on a novel vectorized formulation of the operational neurons.

68, TITLE:      Image Classification in the Dark using Quanta Image Sensors
http://arxiv.org/abs/2006.02026
AUTHORS:        Abhiram Gnanasambandam ; Stanley H. Chan
HIGHLIGHT:      In this paper, we present a new low-light image classification solution using Quanta Image Sensors (QIS).

69, TITLE:      Norm-Based Curriculum Learning for Neural Machine Translation
http://arxiv.org/abs/2006.02014
AUTHORS:        Xuebo Liu ; Houtim Lai ; Derek F. Wong ; Lidia S. Chao
COMMENTS:       Accepted to ACL 2020
HIGHLIGHT:      In this paper, we aim to improve the efficiency of training an NMT by introducing a novel norm-based curriculum learning method.



==========Updates to Previous Papers==========
1, TITLE:       Neural Networks with Small Weights and Depth-Separation Barriers
http://arxiv.org/abs/2006.00625
AUTHORS:        Gal Vardi ; Ohad Shamir
HIGHLIGHT:      In this paper, we focus on feedforward ReLU networks, and prove fundamental barriers to proving such results beyond depth $4$, by reduction to open problems and natural-proof barriers in circuit complexity.

2, TITLE:       Sato: Contextual Semantic Type Detection in Tables
http://arxiv.org/abs/1911.06311
AUTHORS:        Dan Zhang ; Yoshihiko Suhara ; Jinfeng Li ; Madelon Hulsebos ; Çağatay Demiralp ; Wang-Chiew Tan
COMMENTS:       VLDB'20
HIGHLIGHT:      We introduce Sato, a hybrid machine learning model to automatically detect the semantic types of columns in tables, exploiting the signals from the context as well as the column values.

3, TITLE:       On the complexity of the clone membership problem
http://arxiv.org/abs/1909.12211
AUTHORS:        Emil Jeřábek
COMMENTS:       29 pages
HIGHLIGHT:      We investigate the complexity of the Boolean clone membership problem (CMP): given a set of Boolean functions $F$ and a Boolean function $f$, determine if $f$ is in the clone generated by $F$, i.e., if it can be expressed by a circuit with $F$-gates.

4, TITLE:       Self-Organization and Artificial Life
http://arxiv.org/abs/1903.07456
AUTHORS:        Carlos Gershenson ; Vito Trianni ; Justin Werfel ; Hiroki Sayama
COMMENTS:       24 pages, 1 figure, 1 table arXiv admin note: substantial text overlap with arXiv:1804.01144
HIGHLIGHT:      In this review, we discuss the fundamental aspects of self-organization and list the main usages within three primary ALife domains, namely "soft" (mathematical/computational modeling), "hard" (physical robots), and "wet" (chemical/biological systems) ALife.

5, TITLE:       Implicitly Defined Layers in Neural Networks
http://arxiv.org/abs/2003.01822
AUTHORS:        Qianggong Zhang ; Yanyang Gu ; Michalkiewicz Mateusz ; Mahsa Baktashmotlagh ; Anders Eriksson
HIGHLIGHT:      In this paper we demonstrate that defining individual layers in a neural network \emph{implicitly} provide much richer representations over the standard explicit one, consequently enabling a vastly broader class of end-to-end trainable architectures.

6, TITLE:       Parabolic Approximation Line Search for DNNs
http://arxiv.org/abs/1903.11991
AUTHORS:        Maximus Mutschler ; Andreas Zell
HIGHLIGHT:      A major challenge in current optimization research for deep learning is to automatically find optimal step sizes for each update step.

7, TITLE:       Data Diversification: An Elegant Strategy For Neural Machine Translation
http://arxiv.org/abs/1911.01986
AUTHORS:        Xuan-Phi Nguyen ; Shafiq Joty ; Wu Kui ; Ai Ti Aw
HIGHLIGHT:      We introduce Data Diversification: a simple strategy to boost neural machine translation (NMT) performance.

8, TITLE:       Feature-map-level Online Adversarial Knowledge Distillation
http://arxiv.org/abs/2002.01775
AUTHORS:        Inseop Chung ; SeongUk Park ; Jangho Kim ; Nojun Kwak
HIGHLIGHT:      Thus in this paper, we propose an online knowledge distillation method that transfers not only the knowledge of the class probabilities but also that of the feature map using the adversarial training framework.

9, TITLE:       Dynamic Term-Modal Logics for First-Order Epistemic Planning
http://arxiv.org/abs/1906.06047
AUTHORS:        Andrés Occhipinti Liberman ; Andreas Achen ; Rasmus Kræmmer Rendsvig
HIGHLIGHT:      The paper introduces an epistemic language with a possible-worlds semantics, followed by novel dynamics given by first-order action models and their execution via product updates.

10, TITLE:      Fusion of Real Time Thermal Image and 1D/2D/3D Depth Laser Readings for Remote Thermal Sensing in Industrial Plants by Means of UAVs and/or Robots
http://arxiv.org/abs/2006.01286
AUTHORS:        Corneliu Arsene
HIGHLIGHT:      This paper presents fast procedures for thermal infrared remote sensing in dark, GPS-denied environments, such as those found in industrial plants such as in High-Voltage Direct Current (HVDC) converter stations.

11, TITLE:      GoEmotions: A Dataset of Fine-Grained Emotions
http://arxiv.org/abs/2005.00547
AUTHORS:        Dorottya Demszky ; Dana Movshovitz-Attias ; Jeongwoo Ko ; Alan Cowen ; Gaurav Nemade ; Sujith Ravi
COMMENTS:       Accepted to ACL 2020
HIGHLIGHT:      We introduce GoEmotions, the largest manually annotated dataset of 58k English Reddit comments, labeled for 27 emotion categories or Neutral.

12, TITLE:      Learning Resilient Behaviors for Navigation Under Uncertainty
http://arxiv.org/abs/1910.09998
AUTHORS:        Tingxiang Fan ; Pinxin Long ; Wenxi Liu ; Jia Pan ; Ruigang Yang ; Dinesh Manocha
COMMENTS:       accepted to ICRA 2020
HIGHLIGHT:      In this paper, we consider the problem that a mobile robot learns adaptive and resilient behaviors for navigating in unseen uncertain environments while avoiding collisions.

13, TITLE:      Memory capacity of neural networks with threshold and ReLU activations
http://arxiv.org/abs/2001.06938
AUTHORS:        Roman Vershynin
COMMENTS:       26 pages. Minor inaccuracies corrected, discussion of prior work expanded
HIGHLIGHT:      Addressing a 1988 open question of Baum, we prove that this phenomenon holds for general multilayered perceptrons, i.e. neural networks with threshold activation functions, or with any mix of threshold and ReLU activations.

14, TITLE:      When2com: Multi-Agent Perception via Communication Graph Grouping
http://arxiv.org/abs/2006.00176
AUTHORS:        Yen-Cheng Liu ; Junjiao Tian ; Nathaniel Glaser ; Zsolt Kira
COMMENTS:       Accepted to CVPR 2020; for the project page, see https://ycliu93.github.io/projects/multi-agent-perception.html
HIGHLIGHT:      In this paper, we address the collaborative perception problem, where one agent is required to perform a perception task and can communicate and share information with other agents on the same task.

15, TITLE:      CDL: Curriculum Dual Learning for Emotion-Controllable Response Generation
http://arxiv.org/abs/2005.00329
AUTHORS:        Lei Shen ; Yang Feng
COMMENTS:       To appear at ACL 2020 (long paper)
HIGHLIGHT:      To alleviate these problems, we propose a novel framework named Curriculum Dual Learning (CDL) which extends the emotion-controllable response generation to a dual task to generate emotional responses and emotional queries alternatively.

16, TITLE:      CausaLM: Causal Model Explanation Through Counterfactual Language Models
http://arxiv.org/abs/2005.13407
AUTHORS:        Amir Feder ; Nadav Oved ; Uri Shalit ; Roi Reichart
COMMENTS:       Our code and data are available at: https://amirfeder.github.io/CausaLM/
HIGHLIGHT:      To bridge that gap, we propose CausaLM, a framework for producing causal model explanations using counterfactual language representation models.

17, TITLE:      Abstractive Text Classification Using Sequence-to-convolution Neural Networks
http://arxiv.org/abs/1805.07745
AUTHORS:        Taehoon Kim ; Jihoon Yang
HIGHLIGHT:      We propose a new deep neural network model and its training scheme for text classification.

18, TITLE:      Layer-Wise Cross-View Decoding for Sequence-to-Sequence Learning
http://arxiv.org/abs/2005.08081
AUTHORS:        Fenglin Liu ; Xuancheng Ren ; Guangxiang Zhao ; Xu Sun ; Liangyou Li
COMMENTS:       Achieve state-of-the-art BLEU scores on WMT14 EN-DE, EN-FR, and IWSLT DE-EN datasets
HIGHLIGHT:      In this work, we propose to encourage the decoder to take the full advantage of the multi-level source representations for layer-wise cross-view decoding.

19, TITLE:      Spitzoid Lesions Diagnosis based on GA feature selection and Random Forest
http://arxiv.org/abs/2003.04745
AUTHORS:        Abir Belaala ; Labib Sadek ; Noureddine Zerhouni ; Christine Devalland
HIGHLIGHT:      Results obtained in this study have potential to open new opportunities in diagnosis of spitzoid lesions.

20, TITLE:      Data Manipulation: Towards Effective Instance Learning for Neural Dialogue Generation via Learning to Augment and Reweight
http://arxiv.org/abs/2004.02594
AUTHORS:        Hengyi Cai ; Hongshen Chen ; Yonghao Song ; Cheng Zhang ; Xiaofang Zhao ; Dawei Yin
COMMENTS:       To appear at ACL 2020 (long paper)
HIGHLIGHT:      In this paper, we propose a data manipulation framework to proactively reshape the data distribution towards reliable samples by augmenting and highlighting effective learning samples as well as reducing the effect of inefficient samples simultaneously.

21, TITLE:      On the Complexity of Exact Pattern Matching in Graphs: Binary Strings and Bounded Degree
http://arxiv.org/abs/1901.05264
AUTHORS:        Massimo Equi ; Roberto Grossi ; Veli Mäkinen
COMMENTS:       Using Lemma 12 and Lemma 13 might to be enough to prove Lemma 14. However, the proof of Lemma 14 is correct if you assume that the graph used in the reduction is a DAG. Hence, since the problem is already quadratic for a DAG and a binary alphabet, it has to be quadratic also for a general graph and a binary alphabet
HIGHLIGHT:      We describe a simple conditional lower bound that, for any constant $\epsilon>0$, an $O(|E|^{1 - \epsilon} \, m)$-time or an $O(|E| \, m^{1 - \epsilon})$-time algorithm for exact pattern matching on graphs, with node labels and patterns drawn from a binary alphabet, cannot be achieved unless the Strong Exponential Time Hypothesis (SETH) is false.

22, TITLE:      AI and Accessibility: A Discussion of Ethical Considerations
http://arxiv.org/abs/1908.08939
AUTHORS:        Meredith Ringel Morris
COMMENTS:       Preprint of a "Viewpoint" column that was published in the Communications of the ACM in May/June 2020
HIGHLIGHT:      Considering the needs of users with disabilities can help technologists identify high-impact challenges whose solutions can advance the state of AI for all users; however, ethical challenges such as inclusivity, bias, privacy, error, expectation setting, simulated data, and social acceptability must be considered.

23, TITLE:      Harnessing adversarial examples with a surprisingly simple defense
http://arxiv.org/abs/2004.13013
AUTHORS:        Ali Borji
HIGHLIGHT:      I introduce a very simple method to defend against adversarial examples.

24, TITLE:      Learning to Accelerate Decomposition for Multi-Directional 3D Printing
http://arxiv.org/abs/2004.03450
AUTHORS:        Chenming Wu ; Yong-Jin Liu ; Charlie C. L. Wang
COMMENTS:       8 pages
HIGHLIGHT:      Recent work proposed a beam-guided search algorithm to find an optimized sequence of plane-clipping, which gives volume decomposition of a given 3D model.

25, TITLE:      Spiking Deep Residual Network
http://arxiv.org/abs/1805.01352
AUTHORS:        Yangfan Hu ; Huajin Tang ; Gang Pan
HIGHLIGHT:      In this paper, we propose an efficient approach to build a spiking version of deep residual network (ResNet).

26, TITLE:      Edge Matching with Inequalities, Triangles, Unknown Shape, and Two Players
http://arxiv.org/abs/2002.03887
AUTHORS:        Jeffrey Bosboom ; Charlotte Chen ; Lily Chung ; Spencer Compton ; Michael Coulombe ; Erik D. Demaine ; Martin L. Demaine ; Ivan Tadeu Ferreira Antunes Filho ; Dylan Hendrickson ; Adam Hesterberg ; Calvin Hsu ; William Hu ; Oliver Korten ; Zhezheng Luo ; Lillian Zhang
COMMENTS:       29 pages, 18 figures. Thorough revisions of Sections 4, 5, and 6/7 (merged)
HIGHLIGHT:      We analyze the computational complexity of several new variants of edge-matching puzzles.

27, TITLE:      Infinity in computable probability
http://arxiv.org/abs/1101.3578
AUTHORS:        Maarten McKubre-Jordens ; Phillip L. Wilson
HIGHLIGHT:      Here we use recursive function theory to examine the popular scenario of an infinite collection of typing monkeys reproducing the works of Shakespeare.

28, TITLE:      Learning Algorithmic Solutions to Symbolic Planning Tasks with a Neural Computer Architecture
http://arxiv.org/abs/1911.00926
AUTHORS:        Daniel Tanneberg ; Elmar Rueckert ; Jan Peters
HIGHLIGHT:      Therefore, we present a novel architecture, based on memory-augmented networks, that is inspired by the von Neumann and Harvard architectures of modern computers.

29, TITLE:      Characterizing the Decision Boundary of Deep Neural Networks
http://arxiv.org/abs/1912.11460
AUTHORS:        Hamid Karimi ; Tyler Derr ; Jiliang Tang
COMMENTS:       Please contact the first author for any issue or the question regarding this paper
HIGHLIGHT:      To achieve this, we propose a novel approach we call Deep Decision boundary Instance Generation (DeepDIG). Then, we introduce a set of important principled characteristics that take advantage of the generated instances near the decision boundary to provide multifaceted understandings of deep neural networks.

30, TITLE:      The NTNU System at the Interspeech 2020 Non-Native Children's Speech ASR Challenge
http://arxiv.org/abs/2005.08433
AUTHORS:        Tien-Hong Lo ; Fu-An Chao ; Shi-Yan Weng ; Berlin Chen
COMMENTS:       Submitted to Interspeech 2020 Special Session: Shared Task on Automatic Speech Recognition for Non-Native Children's Speech
HIGHLIGHT:      This paper describes the NTNU ASR system participating in the Interspeech 2020 Non-Native Children's Speech ASR Challenge supported by the SIG-CHILD group of ISCA.

31, TITLE:      Transforming Multi-Concept Attention into Video Summarization
http://arxiv.org/abs/2006.01410
AUTHORS:        Yen-Ting Liu ; Yu-Jhe Li ; Yu-Chiang Frank Wang
HIGHLIGHT:      In this paper, we propose an novel attention-based framework for video summarization with complex video data.

32, TITLE:      Geometric algorithms for predicting resilience and recovering damage in neural networks
http://arxiv.org/abs/2005.11603
AUTHORS:        Guruprasad Raghavan ; Jiayi Li ; Matt Thomson
COMMENTS:       10 pages and 4 figures
HIGHLIGHT:      In this paper, we establish a mathematical framework to analyze the resilience of artificial neural networks through the lens of differential geometry.

33, TITLE:      Dialogue Coherence Assessment Without Explicit Dialogue Act Labels
http://arxiv.org/abs/1908.08486
AUTHORS:        Mohsen Mesgar ; Sebastian Bücker ; Iryna Gurevych
COMMENTS:       Accepted at ACL 2020
HIGHLIGHT:      We address these issues by introducing a novel approach to dialogue coherence assessment.

34, TITLE:      DeepMNavigate: Deep Reinforced Multi-Robot Navigation Unifying Local & Global Collision Avoidance
http://arxiv.org/abs/1910.09441
AUTHORS:        Qingyang Tan ; Tingxiang Fan ; Jia Pan ; Dinesh Manocha
HIGHLIGHT:      We present a novel algorithm (DeepMNavigate) for global multi-agent navigation in dense scenarios using deep reinforcement learning (DRL).

35, TITLE:      Named Entity Recognition as Dependency Parsing
http://arxiv.org/abs/2005.07150
AUTHORS:        Juntao Yu ; Bernd Bohnet ; Massimo Poesio
COMMENTS:       Accepted by ACL 2020
HIGHLIGHT:      In this paper, we use ideas from graph-based dependency parsing to provide our model a global view on the input via a biaffine model (Dozat and Manning, 2017).

36, TITLE:      Plexus Convolutional Neural Network (PlexusNet): A novel neural network architecture for histologic image analysis
http://arxiv.org/abs/1908.09067
AUTHORS:        Okyaz Eminaga ; Mahmoud Abbas ; Christian Kunder ; Andreas M. Loening ; Jeanne Shen ; James D. Brooks ; Curtis P. Langlotz ; Daniel L. Rubin
HIGHLIGHT:      Given these limitations, we introduced a novel architecture (termed PlexusNet).

37, TITLE:      Adaptive and Azimuth-Aware Fusion Network of Multimodal Local Features for 3D Object Detection
http://arxiv.org/abs/1910.04392
AUTHORS:        Yonglin Tian ; Kunfeng Wang ; Yuang Wang ; Yulin Tian ; Zilei Wang ; Fei-Yue Wang
COMMENTS:       Accepted by Neurocomputing
HIGHLIGHT:      This paper focuses on the construction of stronger local features and the effective fusion of image and LiDAR data.

38, TITLE:      Differentiate Everything with a Reversible Domain-Specific Language
http://arxiv.org/abs/2003.04617
AUTHORS:        Jin-Guo Liu ; Taine Zhao
COMMENTS:       Github: https://github.com/GiggleLiu/NiLang.jl
HIGHLIGHT:      This paper answers the question that how practical it is to implement a machine instruction level reverse mode AD in a reversible programming language.

39, TITLE:      Engineering AI Systems: A Research Agenda
http://arxiv.org/abs/2001.07522
AUTHORS:        Jan Bosch ; Ivica Crnkovic ; Helena Holmström Olsson
COMMENTS:       8 pages, 4 figures
HIGHLIGHT:      In this paper, we provide a conceptualization of the typical evolution patterns that companies experience when employing ML as well as an overview of the key problems experienced by the companies that we have studied.

40, TITLE:      A Schematic Definition of Quantum Polynomial Time Computability
http://arxiv.org/abs/1802.02336
AUTHORS:        Tomoyuki Yamakami
COMMENTS:       A4, 10pt, pp.29. This is a complete and corrected version of an extended abstract that appeared in the Proceedings of the 9th Workshop on Non-Classical Models of Automata and Applications (NCMA 2017), Prague, Czech Republic, August 17-18, 2017, the Austrian Computer Society, 2017
HIGHLIGHT:      For quantum functions mapping finite-dimensional Hilbert spaces to themselves, we present such a schematic definition, composed of a small set of initial quantum functions and a few construction rules that dictate how to build a new quantum function from the existing ones.

41, TITLE:      Neural Analogical Matching
http://arxiv.org/abs/2004.03573
AUTHORS:        Maxwell Crouse ; Constantine Nakos ; Ibrahim Abdelaziz ; Kenneth Forbus
COMMENTS:       added link to open source code
HIGHLIGHT:      As part of the first steps towards such an integration, we introduce the Analogical Matching Network: a neural architecture that learns to produce analogies between structured, symbolic representations that are largely consistent with the principles of Structure-Mapping Theory.

42, TITLE:      Rethinking Dialogue State Tracking with Reasoning
http://arxiv.org/abs/2005.13129
AUTHORS:        Lizi Liao ; Yunshan Ma ; Wenqiang Lei ; Tat-Seng Chua
COMMENTS:       further modification needed
HIGHLIGHT:      This paper proposes to track dialogue states gradually with reasoning over dialogue turns with the help of the back-end data.

43, TITLE:      Explainable Artificial Intelligence: a Systematic Review
http://arxiv.org/abs/2006.00093
AUTHORS:        Giulia Vilone ; Luca Longo
COMMENTS:       78 pages, 18 figures, journal paper to be submitted to ACM Computing Surveys
HIGHLIGHT:      This systematic review contributes to the body of knowledge by clustering these methods with a hierarchical classification system with four main clusters: review articles, theories and notions, methods and their evaluation.

44, TITLE:      Differentiated Backprojection Domain Deep Learning for Conebeam Artifact Removal
http://arxiv.org/abs/1906.06854
AUTHORS:        Yoseob Han ; Junyoung Kim ; Jong Chul Ye
COMMENTS:       This paper is accepted for IEEE Trans. Medical Imaging
HIGHLIGHT:      In this paper, we develop a novel deep learning approach for accurate conebeam artifact removal.