机器学习每日论文速递[09.24]
同步公众号(arXiv每日论文速递),欢迎关注,感谢支持哦~
cs.LG 方向,今日共计86篇
【1】 Improving Generative Visual Dialog by Answering Diverse Questions
标题:通过回答不同的问题改进生成性视觉对话
作者: Vishvak Murahari, Abhishek Das
链接:https://arxiv.org/abs/1909.10470
【2】 Model-Agnostic Linear Competitors -- When Interpretable Models Compete and Collaborate with Black-Box Models
标题:与模型无关的线性竞争对手-当可解释的模型与黑盒模型竞争和协作时
作者: Hassan Rafique, Qihang Lin
链接:https://arxiv.org/abs/1909.10467
【3】 PAC Reinforcement Learning without Real-World Feedback
标题:无真实反馈的PAC强化学习
作者: Yuren Zhong, Clayton Scott
链接:https://arxiv.org/abs/1909.10449
【4】 On Model Stability as a Function of Random Seed
标题:关于模型稳定性作为随机种子的函数
作者: Pranava Madhyastha, Rishabh Jain
备注:v1; Accepted for publication at CoNLL 2019
链接:https://arxiv.org/abs/1909.10447
【5】 Scalable Kernel Learning via the Discriminant Information
标题:基于鉴别信息的可伸缩核学习
作者: Mert Al, Sun-Yuan Kung
链接:https://arxiv.org/abs/1909.10432
【6】 Class-dependent Compression of Deep Neural Networks
标题:深度神经网络的类相关压缩
作者: Rahim Entezari, Olga Saukh
链接:https://arxiv.org/abs/1909.10364
【7】 Heterogeneous Graph Convolutional Networks for Temporal Community Detection
标题:用于时态社区检测的异构图卷积网络
作者: Yaping Zheng, Di Wang
链接:https://arxiv.org/abs/1909.10248
【8】 Verified Uncertainty Calibration
标题:验证的不确定度校准
作者: Ananya Kumar, Tengyu Ma
备注:Accepted as a spotlight to NeurIPS 2019, original title was "Variance Reduced Uncertainty Calibration"
链接:https://arxiv.org/abs/1909.10155
【9】 Deep Universal Graph Embedding Neural Network
标题:深度泛图嵌入神经网络
作者: Saurabh Verma, Zhi-Li Zhang
链接:https://arxiv.org/abs/1909.10086
【10】 Analyzing Recurrent Neural Network by Probabilistic Abstraction
标题:用概率抽象法分析递归神经网络
作者: Guoliang Dong, Jin Song Dong
链接:https://arxiv.org/abs/1909.10023
【11】 Multi-task Learning and Catastrophic Forgetting in Continual Reinforcement Learning
标题:连续强化学习中的多任务学习与灾难性遗忘
作者: João Ribeiro, João Dias
链接:https://arxiv.org/abs/1909.10008
【12】 Minimal Learning Machine: Theoretical Results and Clustering-Based Reference Point Selection
标题:最小学习机:理论结果和基于聚类的参考点选择
作者: Joonas Hämäläinen, João P. P. Gomes
链接:https://arxiv.org/abs/1909.09978
【13】 LoGANv2: Conditional Style-Based Logo Generation with Generative Adversarial Networks
标题:LoGANv2:具有生成性对抗网络的基于条件样式的Logo生成
作者: Cedric Oeldorf, Gerasimos Spanakis
备注:accepted for poster presentation at ICMLA 2019, data+code available: this https URL
链接:https://arxiv.org/abs/1909.09974
【14】 Classification in asymmetric spaces via sample compression
标题:基于样本压缩的非对称空间分类
作者: Lee-Ad Gottlieb, Shira Ozeri
链接:https://arxiv.org/abs/1909.09969
【15】 HAWKEYE: Adversarial Example Detector for Deep Neural Networks
标题:Hawkeye:深度神经网络的敌对示例检测器
作者: Jinkyu Koo, Saurabh Bagchi
链接:https://arxiv.org/abs/1909.09938
【16】 MaLTESE: Large-Scale Simulation-Driven Machine Learning for Transient Driving Cycles
标题:马耳他:用于瞬态驱动周期的大规模仿真驱动机器学习
作者: Shashi M. Aithal, Prasanna Balaprakash
链接:https://arxiv.org/abs/1909.09929
【17】 Deep Reinforcement Learning with Modulated Hebbian plus Q Network Architecture
标题:使用调制Hebbian plus Q网络结构的深度强化学习
作者: Pawel Ladosz, Andrea Soltoggio
链接:https://arxiv.org/abs/1909.09902
【18】 Uncertainty Quantification with Statistical Guarantees in End-to-End Autonomous Driving Control
标题:端到端自主驾驶控制中具有统计保证的不确定性量化
作者: Rhiannon Michelmore, Marta Kwiatkowska
备注:7 pages, 3 figures, submitted to ICRA 2020
链接:https://arxiv.org/abs/1909.09884
【19】 Deep Message Passing on Sets
标题:集合上的深度消息传递
作者: Yifeng Shi, Marc Niethammer
链接:https://arxiv.org/abs/1909.09877
【20】 On the Importance of Delexicalization for Fact Verification
标题:论事实核查中去词典化的重要性
作者: Sandeep Suntwal, Mihai Surdeanu
备注:submitted as an extended abstract to blackboxnlp workshop under acl 2019
链接:https://arxiv.org/abs/1909.09868
【21】 Single Class Universum-SVM
标题:单类Universum-SVM
作者: Sauptik Dhar, Vladimir Cherkassky
链接:https://arxiv.org/abs/1909.09862
【22】 An Investigation of Quantum Deep Clustering Framework with Quantum Deep SVM & Convolutional Neural Network Feature Extractor
标题:基于量子深度支持向量机和卷积神经网络特征提取的量子深度聚类框架研究
作者: Arit Kumar Bishwas, Vasile Palade
链接:https://arxiv.org/abs/1909.09852
【23】 Using theoretical ROC curves for analysing machine learning binary classifiers
标题:利用理论ROC曲线分析机器学习二元分类器
作者: Luma Omar, Ioannis Ivrissimtzis
链接:https://arxiv.org/abs/1909.09816
【24】 Positive-Unlabeled Compression on the Cloud
标题:云上的正无标签压缩
作者: Yixing Xu, Chang Xu
链接:https://arxiv.org/abs/1909.09757
【25】 Scale MLPerf-0.6 models on Google TPU-v3 Pods
标题:在Google TPU-v3 Pods上缩放MLPerf-0.6型号
作者: Sameer Kumar, Zongwei Zhou
链接:https://arxiv.org/abs/1909.09756
【26】 Learning an Adaptive Learning Rate Schedule
标题:学习自适应学习率计划
作者: Zhen Xu, Luke Metz
链接:https://arxiv.org/abs/1909.09712
【27】 Do Compressed Representations Generalize Better?
标题:压缩表示法能更好地概括吗?
作者: Hassan Hafez-Kolahi, Mahdiyeh Soleymani-Baghshah
链接:https://arxiv.org/abs/1909.09706
【28】 A Layered Architecture for Active Perception: Image Classification using Deep Reinforcement Learning
标题:一种用于主动感知的分层体系结构:使用深度强化学习的图像分类
作者: Hossein K. Mousavi, Nader Motee
备注:Submitted to ICRA-2020
链接:https://arxiv.org/abs/1909.09705
【29】 Online Hierarchical Clustering Approximations
标题:在线层次聚类近似
作者: Aditya Krishna Menon, Sanjiv Kumar
链接:https://arxiv.org/abs/1909.09667
【30】 Understanding and Robustifying Differentiable Architecture Search
标题:对可区分体系结构搜索的理解和推广化
作者: Arber Zela, Frank Hutter
链接:https://arxiv.org/abs/1909.09656
【31】 Accident Risk Prediction based on Heterogeneous Sparse Data: New Dataset and Insights
标题:基于异构稀疏数据的事故风险预测:新的数据集和见解
作者: Sobhan Moosavi, Rajiv Ramnath
备注:In Proceedings of the 27th ACM SIGSPATIAL, International Conference on Advances in Geographic Information Systems (2019). arXiv admin note: substantial text overlap with arXiv:1906.05409
链接:https://arxiv.org/abs/1909.09638
【32】 Learning Dense Representations for Entity Retrieval
标题:实体检索中的稠密表示学习
作者: Daniel Gillick, Diego Garcia-Olano
备注:CoNLL 2019
链接:https://arxiv.org/abs/1909.10506
【33】 Constrained Attractor Selection Using Deep Reinforcement Learning
标题:基于深度强化学习的约束吸引子选择
作者: Xue-She Wang, Brian P. Mann
链接:https://arxiv.org/abs/1909.10500
【34】 Reduced network extremal ensemble learning (RenEEL) scheme for community detection in complex networks
标题:用于复杂网络中社区检测的简化网络极值集成学习(RenEEL)方案
作者: Jiahao Guo, Kevin E. Bassler
链接:https://arxiv.org/abs/1909.10491
【35】 Adversarial Examples for Deep Learning Cyber Security Analytics
标题:深度学习网络安全分析的对抗性示例
作者: Alesia Chernikova, Alina Oprea
链接:https://arxiv.org/abs/1909.10480
【36】 Community Detection and Improved Detectability in Multiplex Networks
标题:复用网络中的社区检测和改进的可检测性
作者: Yuming Huang, Liyi Dai
链接:https://arxiv.org/abs/1909.10477
【37】 Research Directions in Democratizing Innovation through Design Automation, One-Click Manufacturing Services and Intelligent Machines
标题:通过设计自动化、一键式制造服务和智能机器实现创新民主化的研究方向
作者: Binil Starly, Paul Cohen
链接:https://arxiv.org/abs/1909.10476
【38】 Predicting AC Optimal Power Flows: Combining Deep Learning and Lagrangian Dual Methods
标题:预测交流最优潮流:结合深度学习和拉格朗日对偶方法
作者: Ferdinando Fioretto, Pascal Van Hentenryck
链接:https://arxiv.org/abs/1909.10461
【39】 Necessary and Sufficient Conditions for Adaptive, Mirror, and Standard Gradient Methods
标题:自适应、镜像和标准梯度法的充要条件
作者: Daniel Levy, John C. Duchi
备注:23 pages. To appear at NeurIPS 2019
链接:https://arxiv.org/abs/1909.10455
【40】 CochleaNet: A Robust Language-independent Audio-Visual Model for Speech Enhancement
标题:CochleaNet:一种健壮的与语言无关的语音增强视听模型
作者: Mandar Gogate, Amir Hussain
链接:https://arxiv.org/abs/1909.10407
【41】 Machine Learning Pipelines with Modern Big DataTools for High Energy Physics
标题:具有现代大数据工具的机器学习流水线用于高能物理
作者: Matteo Migliorini, Marco Zanetti
链接:https://arxiv.org/abs/1909.10389
【42】 An Adversarial Approach to Private Flocking in Mobile Robot Teams
标题:移动机器人团队中私人群集的一种对抗性方法
作者: Hehui Zheng (1), (2) Polytechnique Montreal)
链接:https://arxiv.org/abs/1909.10387
【43】 Learning Temporal Attention in Dynamic Graphs with Bilinear Interactions
标题:具有双线性交互作用的动态图的时间注意学习
作者: Boris Knyazev, Graham W. Taylor
链接:https://arxiv.org/abs/1909.10367
【44】 TinyBERT: Distilling BERT for Natural Language Understanding
标题:TinyBERT:用于自然语言理解的BERT提取
作者: Xiaoqi Jiao, Qun Liu
链接:https://arxiv.org/abs/1909.10351
【45】 AHA! an 'Artificial Hippocampal Algorithm' for Episodic Machine Learning
标题:啊哈!一种用于情节机器学习的“人工海马算法”
作者: Gideon Kowadlo, David Rawlinson
链接:https://arxiv.org/abs/1909.10340
【46】 Deep learning architectures for automated image segmentation
标题:用于自动图像分割的深度学习架构
作者: Debleena Sengupta
链接:https://arxiv.org/abs/1909.10333
【47】 How to improve CNN-based 6-DoF camera pose estimation
标题:如何改进基于CNN的6-DoF摄像机姿态估计
作者: Soroush Seifi, Tinne Tuytelaars
链接:https://arxiv.org/abs/1909.10312
【48】 Deep Multi-Facial patches Aggregation Network for Expression Classification from Face Images
标题:用于人脸图像表情分类的深度多面片聚集网络
作者: Amine Djerghri, Alice Othmani
链接:https://arxiv.org/abs/1909.10305
【49】 Where to Look Next: Unsupervised Active Visual Exploration on 360° Input
标题:下一步要看哪里:360°输入上的无监督主动视觉探索
作者: Soroush Seifi, Tinne Tuytelaars
链接:https://arxiv.org/abs/1909.10304
【50】 Conservative set valued fields, automatic differentiation, stochastic gradient method and deep learning
标题:保守集值域,自动微分,随机梯度法和深度学习
作者: Jérôme Bolte, Edouard Pauwels
链接:https://arxiv.org/abs/1909.10300
【51】 Predicting Landscapes from Environmental Conditions Using Generative Networks
标题:利用生成网络从环境条件预测景观
作者: Christian Requena-Mesa, Joachim Denzler
备注:Accepted conference paper at GCPR2019
链接:https://arxiv.org/abs/1909.10296
【52】 Inference of modes for linear stochastic processes
标题:线性随机过程的模态推断
作者: Robert S. MacKay
链接:https://arxiv.org/abs/1909.10247
【53】 Decentralized Markov Chain Gradient Descent
标题:分散马尔可夫链梯度下降
作者: Tao Sun, Dongsheng Li
链接:https://arxiv.org/abs/1909.10238
【54】 Manifold Fitting under Unbounded Noise
标题:无界噪声下的流形拟合
作者: Zhigang Yao, Yuqing Xia
链接:https://arxiv.org/abs/1909.10228
【55】 Deep Convolutions for In-Depth Automated Rock Typing
标题:用于深度自动岩石分类的深层卷积
作者: E.E. Baraboshkin, D.A. Koroteev
链接:https://arxiv.org/abs/1909.10227
【56】 PPINN: Parareal Physics-Informed Neural Network for time-dependent PDEs
标题:PPINN:时变偏微分方程的并行物理信息神经网络
作者: Xuhui Meng, George Em Karniadakis
链接:https://arxiv.org/abs/1909.10145
【57】 LISR: Image Super-resolution under Hardware Constraints
标题:LISR:硬件约束下的图像超分辨率
作者: Pravir Singh Gupta, Gwan Seong Choi
链接:https://arxiv.org/abs/1909.10136
【58】 A generalization of regularized dual averaging and its dynamics
标题:正则化对偶平均的推广及其动力学
作者: Shih-Kang Chao, Guang Cheng
链接:https://arxiv.org/abs/1909.10072
【59】 Cutting the Unnecessary Long Tail: Cost-Effective Big Data Clustering in the Cloud
标题:削减不必要的长尾:云中经济高效的大数据集群
作者: Dongwei Li, Yun Yang
链接:https://arxiv.org/abs/1909.10000
【60】 Adapting Language Models for Non-Parallel Author-Stylized Rewriting
标题:为非并行作者风格化重写适配语言模型
作者: Bakhtiyar Syed, Vasudeva Varma
链接:https://arxiv.org/abs/1909.09962
【61】 Techniques and Applications for Crawling, Ingesting and Analyzing Blockchain Data
标题:抓取、摄取和分析区块链数据的技术和应用
作者: Evan Brinckman, Ian J. Taylor
备注:Manuscript accepted for publication at ICTC 2019 (http://ictc.org)
链接:https://arxiv.org/abs/1909.09925
【62】 Using Chinese Glyphs for Named Entity Recognition
标题:使用中文字形进行命名实体识别
作者: Arijit Sehanobish, Chan Hee Song
链接:https://arxiv.org/abs/1909.09922
【63】 Optimal Learning of Joint Alignments with a Faulty Oracle
标题:具有故障甲骨文的联合比对的最优学习
作者: Kasper Green Larsen, Charalampos E. Tsourakakis
链接:https://arxiv.org/abs/1909.09912
【64】 Deep learning approach to control of prosthetic hands with electromyography signals
标题:肌电信号控制假手的深度学习方法
作者: Mohsen Jafarzadeh, Yonas Tadesse
备注:Conference. Houston, Texas, USA. September, 2019
链接:https://arxiv.org/abs/1909.09910
【65】 Leveraging Human Guidance for Deep Reinforcement Learning Tasks
标题:利用人类指导进行深度强化学习任务
作者: Ruohan Zhang, Peter Stone
备注:Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI 2019)
链接:https://arxiv.org/abs/1909.09906
【66】 A Multi-Strategy Approach to Overcoming Bias in Community Detection Evaluation
标题:一种克服社区检测评估偏差的多策略方法
作者: Jeancarlo Campos Leão (1), (2) Universidade Federal de Minas Gerais)
备注:12 pages, 6 figures, 3 tables. This paper has been submitted to the 34th Brazilian Symposium on Databases, 2019 (SBBD2019)
链接:https://arxiv.org/abs/1909.09903
【67】 Efficient Learning of Distributed Linear-Quadratic Controllers
标题:分布式线性二次型控制器的有效学习
作者: Salar Fattahi, Somayeh Sojoudi
链接:https://arxiv.org/abs/1909.09895
【68】 Sparse Group Lasso: Optimal Sample Complexity, Convergence Rate, and Statistical Inference
标题:稀疏群套索:最优样本复杂性、收敛速度和统计推断
作者: T. Tony Cai, Yuchen Zhou
链接:https://arxiv.org/abs/1909.09851
【69】 Multiagent Evaluation under Incomplete Information
标题:不完全信息下的多智能体评估
作者: Mark Rowland, Remi Munos
链接:https://arxiv.org/abs/1909.09849
【70】 Optimal query complexity for private sequential learning
标题:私有序列学习的最优查询复杂度
作者: Jiaming Xu, Dana Yang
链接:https://arxiv.org/abs/1909.09836
【71】 Automatic Posture and Movement Tracking of Infants with Wearable Movement Sensors
标题:带可穿戴运动传感器的婴儿自动姿势和运动跟踪
作者: Manu Airaksinen, Sampsa Vanhatalo
链接:https://arxiv.org/abs/1909.09823
【72】 ASNI: Adaptive Structured Noise Injection for shallow and deep neural networks
标题:ASNI:用于浅层和深层神经网络的自适应结构化噪声注入
作者: Beyrem Khalfaoui, Jean-Philippe Vert
链接:https://arxiv.org/abs/1909.09819
【73】 Challenges of Privacy-Preserving Machine Learning in IoT
标题:物联网中隐私保护机器学习的挑战
作者: Mengyao Zheng, Peng Cheng
备注:In First International Workshop on Challenges in Artificial Intelligence and Machine Learning (AIChallengeIoT'19) November 10-13, 2019. 7 pages
链接:https://arxiv.org/abs/1909.09804
【74】 Using Statistics to Automate Stochastic Optimization
标题:使用统计方法实现随机优化自动化
作者: Hunter Lang, Lin Xiao
链接:https://arxiv.org/abs/1909.09785
【75】 Teaching Pretrained Models with Commonsense Reasoning: A Preliminary KB-Based Approach
标题:用常识推理教授预先训练的模型:基于知识库的初步方法
作者: Shiyang Li, Dian Yu
链接:https://arxiv.org/abs/1909.09743
【76】 Distributed Parameter Estimation in Randomized One-hidden-layer Neural Networks
标题:随机化单隐层神经网络的分布参数估计
作者: Yinsong Wang, Shahin Shahrampour
链接:https://arxiv.org/abs/1909.09736
【77】 COPYCAT: Practical Adversarial Attacks on Visualization-Based Malware Detection
标题:Copycat:对基于可视化的恶意软件检测的实际对抗性攻击
作者: Aminollah Khormali, Aziz Mohaisen
链接:https://arxiv.org/abs/1909.09735
【78】 Particle Smoothing Variational Objectives
标题:粒子平滑变分目标
作者: Antonio Khalil Moretti, Itsik Pe'er
链接:https://arxiv.org/abs/1909.09734
【79】 Safer End-to-End Autonomous Driving via Conditional Imitation Learning and Command Augmentation
标题:通过条件模仿学习和命令增强实现更安全的端到端自主驾驶
作者: Renhao Wang, Frank Wood
备注:Submitted to the 2020 International Conference on Robotics and Automation
链接:https://arxiv.org/abs/1909.09721
【80】 Measuring Domain Portability and ErrorPropagation in Biomedical QA
标题:生物医学QA中测量域的可移植性和误差传播
作者: Stefan Hosein, Ryan McDonal
链接:https://arxiv.org/abs/1909.09704
【81】 Using Clinical Notes with Time Series Data for ICU Management
标题:使用带有时间序列数据的临床记录进行ICU管理
作者: Swaraj Khadanga, Jaideep Srivastava
备注:Accepted at EMNLP 2019
链接:https://arxiv.org/abs/1909.09702
【82】 Induction and Reference of Entities in a Visual Story
标题:视觉故事中实体的归纳和引用
作者: Ruo-Ping Dong, Alan W Black
链接:https://arxiv.org/abs/1909.09699
【83】 NSURL-2019 Shared Task 8: Semantic Question Similarity in Arabic
标题:NSURL-2019共享任务8:阿拉伯语中的语义问题相似性
作者: Haitham Seelawi, Hussein T. Al-Natsheh
链接:https://arxiv.org/abs/1909.09691
【84】 A Deep Learning-Based Approach for Measuring the Domain Similarity of Persian Texts
标题:一种基于深度学习的波斯文本领域相似度度量方法
作者: Hossein Keshavarz, Mohammad Izadi
链接:https://arxiv.org/abs/1909.09690
【85】 What are Neural Networks made of?
标题:神经网络是由什么组成的?
作者: Rene Schaub
链接:https://arxiv.org/abs/1909.09588
【86】 Deep neural network approximations for Monte Carlo algorithms
标题:蒙特卡罗算法的深层神经网络逼近
作者: Philipp Grohs, Diyora Salimova
链接:https://arxiv.org/abs/1908.10828
机器翻译,仅供参考