机器学习每日论文速递[07.12]

同步wx订阅号(arXiv每日论文速递),支持后台回复'search 关键词'查询相关的最新论文。有些许帮助的话,麻烦关注一下哦(* ̄rǒ ̄)


cs.LG 方向,今日共计71篇

[cs.LG]:

【1】 Provably Efficient Reinforcement Learning with Linear Function Approximation
标题:线性函数逼近的可证明有效的强化学习
作者: Chi Jin, Michael I. Jordan
链接:arxiv.org/abs/1907.0538

【2】 Performance Boundary Identification for the Evaluation of Automated Vehicles using Gaussian Process Classification
标题:基于高斯过程分类的自动车辆性能边界识别
作者: Felix Batsch, Anthony Baxendale
备注:6 pages, 5 figures, accepted at 2019 IEEE Intelligent Transportation Systems Conference - ITSC 2019, Auckland, New Zealand, October 2019
链接:arxiv.org/abs/1907.0536

【3】 Time2Vec: Learning a Vector Representation of Time
标题:Time2Vec:学习时间的向量表示
作者: Seyed Mehran Kazemi, Marcus Brubaker
链接:arxiv.org/abs/1907.0532

【4】 Beyond Imitation: Generative and Variational Choreography via Machine Learning
标题:超越模仿:通过机器学习的生成性和变化性编排
作者: Mariel Pettee, Ilya Vidrin
备注:8 pages, 11 figures, presented at the 10th International Conference on Computational Creativity (ICCC 2019)
链接:arxiv.org/abs/1907.0529

【5】 Perturbation theory approach to study the latent space degeneracy of Variational Autoencoders
标题:研究变分自动编码器潜在空间简并性的微扰理论方法
作者: Helena Andrés-Terré, Pietro Lió
链接:arxiv.org/abs/1907.0526

【6】 Time series cluster kernels to exploit informative missingness and incomplete label information
标题:时间序列聚类核利用信息性缺失和不完整的标签信息
作者: Karl Øyvind Mikalsen, Robert Jenssen
备注:arXiv admin note: text overlap with arXiv:1803.07879
链接:arxiv.org/abs/1907.0525

【7】 Large Memory Layers with Product Keys
标题:具有产品密钥的大内存层
作者: Guillaume Lample, Hervé Jégou
链接:arxiv.org/abs/1907.0524

【8】 Identifying Linear Models in Multi-Resolution Population Data using Minimum Description Length Principle to Predict Household Income
标题:用最小描述长度原理识别多分辨率人口数据中的线性模型预测家庭收入
作者: Chainarong Amornbunchornvej, Suttipong Thajchayapong
链接:arxiv.org/abs/1907.0523

【9】 Variance-Based Risk Estimations in Markov Processes via Transformation with State Lumping
标题:基于状态集总变换的马尔可夫过程中基于方差的风险估计
作者: Shuai Ma, Jia Yuan Yu
备注:7 pages, 7 figures, SMC 2019 accepted. arXiv admin note: text overlap with arXiv:1907.04269
链接:arxiv.org/abs/1907.0523

【10】 Eigen Artificial Neural Networks
标题:特征型人工神经网络
作者: Francisco Yepes Barrera
链接:arxiv.org/abs/1907.0520

【11】 Fairness without Regret
标题:公平无悔
作者: Marcus Hutter
链接:arxiv.org/abs/1907.0515

【12】 Amplifying Rényi Differential Privacy via Shuffling
标题:通过混洗放大Rényi差分隐私
作者: Eloïse Berthier, Sai Praneeth Karimireddy
链接:arxiv.org/abs/1907.0515

【13】 Forecasting remaining useful life: Interpretable deep learning approach via variational Bayesian inferences
标题:预测剩余使用寿命:通过变分贝叶斯推断的可解释深度学习方法
作者: Mathias Kraus, Stefan Feuerriegel
链接:arxiv.org/abs/1907.0514

【14】 PreCall: A Visual Interface for Threshold Optimization in ML Model Selection
标题:PreCall:ML模型选择中阈值优化的可视化界面
作者: Christoph Kinkeldey, Aaron Halfaker
备注:HCML Perspectives Workshop at CHI 2019, May 04, 2019, Glasgow
链接:arxiv.org/abs/1907.0513

【15】 Spatiotemporal Local Propagation
标题:时空局部传播
作者: Alessandro Betti, Marco Gori
链接:arxiv.org/abs/1907.0510

【16】 Safe Policy Improvement with Soft Baseline Bootstrapping
标题:使用软基线引导的安全策略改进
作者: Kimia Nadjahi, Rémi Tachet des Combes
备注:Accepted paper at ECML-PKDD2019
链接:arxiv.org/abs/1907.0507

【17】 Profiling based Out-of-core Hybrid Method for Large Neural Networks
标题:基于仿形的大型神经网络核外混合方法
作者: Yuki Ito, Toshio Endo
链接:arxiv.org/abs/1907.0501

【18】 Making AI Forget You: Data Deletion in Machine Learning
标题:让AI忘记你:机器学习中的数据删除
作者: Antonio Ginart, James Zou
链接:arxiv.org/abs/1907.0501

【19】 Understanding the Representation Power of Graph Neural Networks in Learning Graph Topology
标题:了解图神经网络在学习图拓扑中的表现力
作者: Nima Dehmamy, Rose Yu
链接:arxiv.org/abs/1907.0500

【20】 A Model-based Approach for Sample-efficient Multi-task Reinforcement Learning
标题:一种基于模型的有效样本多任务强化学习方法
作者: Nicholas C. Landolfi, Tengyu Ma
链接:arxiv.org/abs/1907.0496

【21】 GraphSAINT: Graph Sampling Based Inductive Learning Method
标题:GraphSAINT:基于图采样的归纳学习方法
作者: Hanqing Zeng, Viktor Prasanna
链接:arxiv.org/abs/1907.0493

【22】 Explaining an increase in predicted risk for clinical alerts
标题:解释临床警报的预测风险增加
作者: Michaela Hardt, Moritz Hardt
链接:arxiv.org/abs/1907.0491

【23】 Interpretable Dynamics Models for Data-Efficient Reinforcement Learning
标题:数据高效强化学习的可解释动力学模型
作者: Markus Kaiser, Carl Henrik Ek
备注:ESANN 2019 proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. Bruges (Belgium), 24-26 April 2019, i6doc.com publ., ISBN 978-287-587-065-0
链接:arxiv.org/abs/1907.0490

【24】 Adversarial Objects Against LiDAR-Based Autonomous Driving Systems
标题:对抗基于LiDAR的自主驾驶系统的对象
作者: Yulong Cao, Bo Li
链接:arxiv.org/abs/1907.0541

【25】 Learning to learn with quantum neural networks via classical neural networks
标题:通过经典神经网络学习量子神经网络
作者: Guillaume Verdon, Masoud Mohseni
链接:arxiv.org/abs/1907.0541

【26】 Change point detection for graphical models in presence of missing values
标题:存在缺失值的图形模型的变化点检测
作者: Malte Londschien, Peter Bühlmann
链接:arxiv.org/abs/1907.0540

【27】 Computational Concentration of Measure: Optimal Bounds, Reductions, and More
标题:测量的计算集中:最优界限、约简和更多
作者: Omid Etesami, Mohammad Mahmoody
链接:arxiv.org/abs/1907.0540

【28】 Single Image Super-Resolution via CNN Architectures and TV-TV Minimization
标题:通过CNN架构和TV-TV最小化实现单图像超分辨率
作者: Marija Vella, João F. C. Mota
备注:Accepted to BMVC 2019
链接:arxiv.org/abs/1907.0538

【29】 Quantum and Classical Algorithms for Approximate Submodular Function Minimization
标题:近似子模函数最小化的量子算法和经典算法
作者: Yassine Hamoudi, Miklos Santha
链接:arxiv.org/abs/1907.0537

【30】 Online Inference and Detection of Curbs in Partially Occluded Scenes with Sparse LIDAR
标题:利用稀疏LIDAR在线推断和检测部分遮挡场景中的限位
作者: Tarlan Suleymanov, Paul Newman
备注:Accepted at the 22nd IEEE Intelligent Transportation Systems Conference (ITSC19), October, 2019, Auckland, New Zealand
链接:arxiv.org/abs/1907.0537

【31】 Warfarin dose estimation on multiple datasets with automated hyperparameter optimisation and a novel software framework
标题:使用自动超参数优化和新的软件框架对多个数据集进行华法林剂量估计
作者: Gianluca Truda, Patrick Marais
链接:arxiv.org/abs/1907.0536

【32】 StrokeSave: A Novel, High-Performance Mobile Application for Stroke Diagnosis using Deep Learning and Computer Vision
标题:StrokeSave:一种利用深度学习和计算机视觉进行中风诊断的新型高性能移动应用
作者: Ankit Gupta
链接:arxiv.org/abs/1907.0535

【33】 To Tune or Not To Tune? How About the Best of Both Worlds?
标题:调不调?那两个世界最好的呢?
作者: Ran Wang, Jupeng Ding
链接:arxiv.org/abs/1907.0533

【34】 Adaptive Margin Ranking Loss for Knowledge Graph Embeddings via a Correntropy Objective Function
标题:基于协熵目标函数的知识图嵌入的自适应边际排序损失
作者: Mojtaba Nayyeri, Jens Lehmann
链接:arxiv.org/abs/1907.0533

【35】 Three algorithms for solving high-dimensional fully-coupled FBSDEs through deep learning
标题:通过深度学习求解高维全耦合FBSDE的三种算法
作者: Shaolin Ji, Xichuan Zhang
链接:arxiv.org/abs/1907.0532

【36】 Low-rank matrix completion and denoising under Poisson noise
标题:泊松噪声下的低秩矩阵完成与去噪
作者: Andrew D. McRae, Mark A. Davenport
链接:arxiv.org/abs/1907.0532

【37】 Adaptive Deep Learning for High Dimensional Hamilton-Jacobi-Bellman Equations
标题:高维Hamilton-Jacobi-Bellman方程的自适应深度学习
作者: Tenavi Nakamura-Zimmerer (1), (2) Naval Postgraduate School)
链接:arxiv.org/abs/1907.0531

【38】 Voxel-FPN: multi-scale voxel feature aggregation in 3D object detection from point clouds
标题:VOXEL-FPN:点云三维物体检测中的多尺度体素特征聚合
作者: Bei Wang, Jiayan Cao
链接:arxiv.org/abs/1907.0528

【39】 A Comparison of Super-Resolution and Nearest Neighbors Interpolation Applied to Object Detection on Satellite Data
标题:超分辨率和最近邻插值在卫星目标检测中的应用比较
作者: Evan Koester, Cem Safak Sahin
链接:arxiv.org/abs/1907.0528

【40】 Image Super-Resolution Using Attention Based DenseNet with Residual Deconvolution
标题:基于剩余反褶积的基于注意力的DenseNet图像超分辨率
作者: Zhuangzi Li
链接:arxiv.org/abs/1907.0528

【41】 City-GAN: Learning architectural styles using a custom Conditional GAN architecture
标题:City-gan:使用自定义条件GaN架构学习建筑风格
作者: Maximilian Bachl, Daniel C. Ferreira
链接:arxiv.org/abs/1907.0528

【42】 Tranquil Clouds: Neural Networks for Learning Temporally Coherent Features in Point Clouds
标题:宁静云:用于学习点云中时间相关特征的神经网络
作者: Lukas Prantl, Stefan Jeschke und Nils Thuerey
链接:arxiv.org/abs/1907.0527

【43】 A General Framework for Complex Network-Based Image Segmentation
标题:一种基于复杂网络的图像分割通用框架
作者: Youssef Mourchid, Hocine Cherifi
链接:arxiv.org/abs/1907.0527

【44】 A Deep Neural Network for Finger Counting and Numerosity Estimation
标题:一种用于手指计数和数量估计的深度神经网络
作者: Leszek Pecyna, Alessandro Di Nuovo
链接:arxiv.org/abs/1907.0527

【45】 Influence of Pointing on Learning to Count: A Neuro-Robotics Model
标题:指点对学习数数的影响:一个神经-机器人模型
作者: Leszek Pecyna, Angelo Cangelosi
备注:8 pages, 5 figures. In Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence (SSCI) (pp. 358-365). IEEE
链接:arxiv.org/abs/1907.0526

【46】 A Deep Reinforcement-Learning-based Driving Policy for Autonomous Road Vehicles
标题:一种基于深度强化学习的自主道路车辆驾驶策略
作者: Konstantinos Makantasis, Ioannis Nikolos
备注:17 pages. arXiv admin note: substantial text overlap with arXiv:1905.09046
链接:arxiv.org/abs/1907.0524

【47】 Gain with no Pain: Efficient Kernel-PCA by Nyström Sampling
标题:无痛收获:Nyström采样的高效核PCA
作者: Nicholas Sterge, Alessandro Rudi
备注:19 pages, 2 figures
链接:arxiv.org/abs/1907.0522

【48】 retina-VAE: Variationally Decoding the Spectrum of Macular Disease
标题:视网膜-VAE:黄斑疾病光谱的变异解码
作者: Stephen G. Odaibo
链接:arxiv.org/abs/1907.0519

【49】 A Neural Architecture for Designing Truthful and Efficient Auctions
标题:设计真实有效的拍卖的神经结构
作者: Andrea Tacchetti, Yoram Bachrach
链接:arxiv.org/abs/1907.0518

【50】 Privileged Features Distillation for E-Commerce Recommendations
标题:用于电子商务推荐的特权功能蒸馏
作者: Chen Xu, Wenwu Ou
链接:arxiv.org/abs/1907.0517

【51】 Disease classification of macular Optical Coherence Tomography scans using deep learning software: validation on independent, multi-centre data
标题:使用深度学习软件对黄斑光学相干断层扫描的疾病分类:对独立的多中心数据的验证
作者: Kanwal K. Bhatia, Nicolas Jaccard
链接:arxiv.org/abs/1907.0516

【52】 Deep Active Learning for Axon-Myelin Segmentation on Histology Data
标题:基于组织学数据的轴突-髓鞘分割的深度主动学习
作者: Melanie Lubrano di Scandalea, Julien Cohen-Adad
链接:arxiv.org/abs/1907.0514

【53】 Kernel Trajectory Maps for Multi-Modal Probabilistic Motion Prediction
标题:多模态概率运动预测的核轨迹图
作者: Weiming Zhi, Fabio Ramos
链接:arxiv.org/abs/1907.0512

【54】 Machine Learning Kernel Method from a Quantum Generative Model
标题:量子生成模型的机器学习核方法
作者: Przemysław Sadowski
链接:arxiv.org/abs/1907.0510

【55】 Analysis of Ward's Method
标题:Ward方法分析
作者: Anna Großwendt, Melanie Schmidt
备注:appeared at SODA 2019
链接:arxiv.org/abs/1907.0509

【56】 Neural Network-based Equalizer by Utilizing Coding Gain in Advance
标题:提前利用编码增益的神经网络均衡器
作者: Chieh-Fang Teng, An-Yeu Wu
备注:5 pages, 4 figures, submitted to 2019 Seventh IEEE Global Conference on Signal and Information Processing
链接:arxiv.org/abs/1907.0498

【57】 Diverse Trajectory Forecasting with Determinantal Point Processes
标题:基于行列点过程的多样化轨迹预测
作者: Ye Yuan, Kris Kitani
链接:arxiv.org/abs/1907.0496

【58】 Can Unconditional Language Models Recover Arbitrary Sentences?
标题:无条件语言模型可以恢复任意句子吗?
作者: Nishant Subramani, Kyunghyun Cho
备注:Under Review at NeurIPS 2019
链接:arxiv.org/abs/1907.0494

【59】 Bag-of-Audio-Words based on Autoencoder Codebook for Continuous Emotion Prediction
标题:基于AUTOCODER码本的连续情感预测的音频包(Bag-of-Audio-Words)
作者: Mohammed Senoussaoui, Alessandro Lameiras Koerich
链接:arxiv.org/abs/1907.0492

【60】 Speech bandwidth extension with WaveNet
标题:利用WaveNet扩展语音带宽
作者: Archit Gupta, Thomas C. Walters
链接:arxiv.org/abs/1907.0492

【61】 Infer Implicit Contexts in Real-time Online-to-Offline Recommendation
标题:实时在线到离线推荐中的隐含上下文推断
作者: Xichen Ding, Dan Shen
备注:9 pages,KDD,KDD2019
链接:arxiv.org/abs/1907.0492

【62】 Interactive Topic Modeling with Anchor Words
标题:基于锚词的交互式主题建模
作者: Sanjoy Dasgupta, Christopher Tosh
备注:presented at 2019 ICML Workshop on Human in the Loop Learning (HILL 2019), Long Beach, USA
链接:arxiv.org/abs/1907.0491

【63】 Listen, Attend, Spell and Adapt: Speaker Adapted Sequence-to-Sequence ASR
标题:聆听、参加、拼写和改编:扬声器适配序列到序列ASR
作者: Felix Weninger, Puming Zhan
备注:To appear in INTERSPEECH 2019
链接:arxiv.org/abs/1907.0491

【64】 Prediction of Compression Index of Fine-Grained Soils Using a Gene Expression Programming Model
标题:用基因表达式编程模型预测细粒土壤压缩指数
作者: Danial Mohammadzadeh, Joseph H. M. Tah
链接:arxiv.org/abs/1907.0491

【65】 Topic Modeling in Embedding Spaces
标题:嵌入空间中的主题建模
作者: Adji B. Dieng, David M. Blei
链接:arxiv.org/abs/1907.0490

【66】 Development of email classifier in Brazilian Portuguese using feature selection for automatic response
标题:使用自动回复的特征选择开发巴西葡萄牙语的电子邮件分类器
作者: Rogerio Bonatti, Arthur Gola de Paula
链接:arxiv.org/abs/1907.0490

【67】 Super-resolution meets machine learning: approximation of measures
标题:超分辨率满足机器学习:测度的近似
作者: H. N. Mhaskar
链接:arxiv.org/abs/1907.0489

【68】 Productization Challenges of Contextual Multi-Armed Bandits
标题:情境多武装匪徒的产品化挑战
作者: David Abensur, Ido Tamir
链接:arxiv.org/abs/1907.0488

【69】 LakhNES: Improving multi-instrumental music generation with cross-domain pre-training
标题:LakhNES:通过跨域预培训改进多乐器音乐生成
作者: Chris Donahue, Julian McAuley
备注:Published as a conference paper at ISMIR 2019
链接:arxiv.org/abs/1907.0486

【70】 GQA: A New Dataset for Real-World Visual Reasoning and Compositional Question Answering
标题:GQA:一种新的用于现实世界视觉推理和组件式问答的数据集
作者: Drew A. Hudson, Christopher D. Manning
备注:Published as a conference paper at CVPR 2019 (oral)
链接:arxiv.org/abs/1902.0950

【71】 Analysis of Triplet Motifs in Biological Signed Oriented Graphs Suggests a Relationship Between Fine Topology and Function
标题:对生物符号定向图中三重基序的分析表明精细拓扑与功能之间的关系
作者: Alberto Calderone, Gianni Cesareni
链接:arxiv.org/abs/1803.0652

翻译:腾讯翻译君

发布于 2019-07-12 09:36