机器学习论文速递[09.11]

论坛 期权论坛 期权     
arXiv每日论文速递   2019-9-11 11:24   324   0
点击上方蓝字关注公众号,如有些许帮助,最后点个“在看”哦
cs.LG 方向,今日共计73篇


【1】 Towards Understanding the Importance of Shortcut Connections in Residual  Networks
标题:了解残差网络中快捷连接的重要性
作者: Tianyi Liu,  Tuo Zhao
备注:Thirty-third Conference on Neural Information Processing Systems, 2019
链接:https://arxiv.org/abs/1909.04653

【2】 Meta-Learning with Implicit Gradients
标题:具有隐含梯度的元学习
作者: Aravind Rajeswaran,  Sergey Levine
备注:NeurIPS 2019. First two authors contributed equally
链接:https://arxiv.org/abs/1909.04630

【3】 Patient trajectory prediction in the Mimic-III dataset, challenges and  pitfalls
标题:MIMIC-III数据集中的患者轨迹预测,挑战和陷阱
作者: Jose F Rodrigues-Jr,
链接:https://arxiv.org/abs/1909.04605

【4】 Prediction of Overall Survival of Brain Tumor Patients
标题:脑肿瘤患者总生存率的预测
作者: Rupal Agravat,  Mehul S Raval
备注:5 pages, IEEE TENCON 2019
链接:https://arxiv.org/abs/1909.04596

【5】 Efficient nonmyopic Bayesian optimization and quadrature
标题:有效的非近视贝叶斯优化与求积
作者: Shali Jiang,  Roman Garnett
链接:https://arxiv.org/abs/1909.04568

【6】 Differentiable Mask Pruning for Neural Networks
标题:神经网络的微分掩模剪枝
作者: Ramchalam Kinattinkara Ramakrishnan,  Vahid Partovi Nia
链接:https://arxiv.org/abs/1909.04567

【7】 Byzantine-Resilient Stochastic Gradient Descent for Distributed  Learning: A Lipschitz-Inspired Coordinate-wise Median Approach
标题:用于分布式学习的拜占庭弹性随机梯度下降:Lipschitz启发的坐标方向中值方法
作者: Haibo Yang,  Jia Liu
链接:https://arxiv.org/abs/1909.04532

【8】 Skin cancer detection based on deep learning and entropy to detect  outlier samples
标题:基于深度学习和熵的皮肤癌检测
作者: Andre G. C. Pacheco,  Thomas Trappenberg
链接:https://arxiv.org/abs/1909.04525

【9】 A Study of Deep Learning for Network Traffic Data Forecasting
标题:深度学习在网络流量数据预测中的应用研究
作者: Benedikt Pfülb,  Sebastian Rieger
备注:16 pages, 12 figures, 28th International Conference on Artificial Neural Networks (ICANN 2019)
链接:https://arxiv.org/abs/1909.04501

【10】 A Deep Learning Framework for Pricing Financial Instruments
标题:金融工具定价的深度学习框架
作者: Qiong Wu,  Zhenming Liu
链接:https://arxiv.org/abs/1909.04497

【11】 Dropout Induced Noise for Co-Creative GAN Systems
标题:用于共创GaN系统的脱落感生噪声
作者: Sabine Wieluch,  Dr. Friedhelm Schwenker
链接:https://arxiv.org/abs/1909.04474

【12】 Adversarial Orthogonal Regression: Two non-Linear Regressions for Causal  Inference
标题:对抗性正交回归:因果推断的两个非线性回归
作者: M. Reza Heydari,  Kun Zhang
链接:https://arxiv.org/abs/1909.04454

【13】 Learning Priors for Adversarial Autoencoders
标题:对抗性自动编码器的学习先验
作者: Hui-Po Wang,  Wei-Jan Ko
备注:Accepted by APSIPA ASC, 2018
链接:https://arxiv.org/abs/1909.04443

【14】 The Prevalence of Errors in Machine Learning Experiments
标题:机器学习实验中错误的盛行
作者: Martin Shepperd,  Leila Yousefi
备注:20th International Conference on Intelligent Data Engineering and Automated Learning (IDEAL), 14--16 November 2019
链接:https://arxiv.org/abs/1909.04436

【15】 Privacy-Preserving Bandits
标题:保护隐私的土匪
作者: Mohammad Malekzadeh,  Ben Livshits
链接:https://arxiv.org/abs/1909.04421

【16】 Compositional Generalization in Image Captioning
标题:图像字幕中的构图概括
作者: Mitja Nikolaus,  Desmond Elliott
备注:To appear at CoNLL 2019, EMNLP
链接:https://arxiv.org/abs/1909.04402

【17】 Classifying the Valence of Autobiographical Memories from fMRI Data
标题:从fMRI数据中分类自传体记忆的价态
作者: Alex Frid,  Norberto Eiji Nawa
链接:https://arxiv.org/abs/1909.04390

【18】 Wasserstein Collaborative Filtering for Item Cold-start Recommendation
标题:用于项目冷启动推荐的Wasserstein协作过滤
作者: Yitong Meng,  Weiwen Liu
链接:https://arxiv.org/abs/1909.04266

【19】 Temporal Network Embedding with Micro- and Macro-dynamics
标题:具有微观和宏观动力学的时态网络嵌入
作者: Yuanfu Lu,  Yanfang Ye
备注:CIKM2019
链接:https://arxiv.org/abs/1909.04246

【20】 Neural reparameterization improves structural optimization
标题:神经重新参数化改进了结构优化
作者: Stephan Hoyer,  Sam Greydanus
链接:https://arxiv.org/abs/1909.04240

【21】 Multi-Step Greedy and Approximate Real Time Dynamic Programming
标题:多步贪婪近似实时动态规划
作者: Yonathan Efroni,  Shie Mannor
链接:https://arxiv.org/abs/1909.04236

【22】 Swapped Face Detection using Deep Learning and Subjective Assessment
标题:基于深度学习和主观评价的互换人脸检测
作者: Xinyi Ding,  Michael Hahsler
链接:https://arxiv.org/abs/1909.04217

【23】 Novel diffusion-derived distance measures for graphs
标题:图的一种新的扩散导出距离测度
作者: C.B. Scott,  Eric Mjolsness
链接:https://arxiv.org/abs/1909.04203

【24】 Augmenting Monte Carlo Dropout Classification Models with Unsupervised  Learning Tasks for Detecting and Diagnosing Out-of-Distribution Faults
标题:用无监督学习任务增强Monte Carlo Dropout分类模型以检测和诊断失配故障
作者: Baihong Jin,  Alberto Sangiovanni-Vincentelli
链接:https://arxiv.org/abs/1909.04202

【25】 NormLime: A New Feature Importance Metric for Explaining Deep Neural  Networks
标题:NormLime:一种新的用于解释深度神经网络的特征重要性度量
作者: Isaac Ahern,  Jun Huan
链接:https://arxiv.org/abs/1909.04200

【26】 Option Encoder: A Framework for Discovering a Policy Basis in  Reinforcement Learning
标题:Option Encoder:一个在强化学习中发现策略基础的框架
作者: Arjun Manoharan,  Balaraman Ravindran
链接:https://arxiv.org/abs/1909.04134

【27】 AC-Teach: A Bayesian Actor-Critic Method for Policy Learning with an  Ensemble of Suboptimal Teachers
标题:AC-Teach:一种贝叶斯角色-评判法在次优教师群体中的政策学习
作者: Andrey Kurenkov,  Animesh Garg
链接:https://arxiv.org/abs/1909.04121

【28】 A Classification Methodology based on Subspace Graphs Learning
标题:基于子空间图学习的分类方法
作者: Riccardo La Grassa,  Dimitri Ognibene
链接:https://arxiv.org/abs/1909.04078

【29】 Adversarial Robustness Against the Union of Multiple Perturbation Models
标题:多扰动模型并集的对抗性鲁棒性
作者: Pratyush Maini,  J. Zico Kolter
链接:https://arxiv.org/abs/1909.04068

【30】 Exploratory Combinatorial Optimization with Reinforcement Learning
标题:基于强化学习的探索性组合优化
作者: Thomas D. Barrett,  A. I. Lvovsky
链接:https://arxiv.org/abs/1909.04063

【31】 A Flexible Framework for Anomaly Detection via Dimensionality Reduction
标题:一种灵活的降维异常检测框架
作者: Alireza Vafaei Sadr,  Martin Kunz
链接:https://arxiv.org/abs/1909.04060

【32】 Deep Learning for Automated Classification and Characterization of  Amorphous Materials
标题:用于非晶态材料自动分类和表征的深度学习
作者: Kirk Swanson,  Risi Kondor
链接:https://arxiv.org/abs/1909.04648

【33】 Representation of Constituents in Neural Language Models: Coordination  Phrase as a Case Study
标题:神经语言模型中成分的表示:以协调短语为例
作者: Aixiu An,  Roger Levy
备注:To appear at EMNLP 2019
链接:https://arxiv.org/abs/1909.04625

【34】 Discovery of Useful Questions as Auxiliary Tasks
标题:发现有用的问题作为辅助任务
作者: Vivek Veeriah,  Satinder Singh
链接:https://arxiv.org/abs/1909.04607

【35】 Road Mapping In LiDAR Images Using A Joint-Task Dense Dilated  Convolutions Merging Network
标题:使用联合任务密集扩张卷积合并网络的LiDAR图像中的道路映射
作者: Qinghui Liu,  Arnt-Brre Salberg
备注:IGARSS 2019. arXiv admin note: text overlap with arXiv:1908.11799
链接:https://arxiv.org/abs/1909.04588

【36】 Scalable Structure Learning of Continuous-Time Bayesian Networks from  Incomplete Data
标题:不完全数据下连续时间贝叶斯网络的可伸缩结构学习
作者: Dominik Linzner,  Heinz Koeppl
备注:Accepted at NeurIPS2019
链接:https://arxiv.org/abs/1909.04570

【37】 PREMA: A Predictive Multi-task Scheduling Algorithm For Preemptible  Neural Processing Units
标题:Prema:一种可抢占神经处理单元的预测多任务调度算法
作者: Yujeong Choi,  Minsoo Rhu
链接:https://arxiv.org/abs/1909.04548

【38】 DeepPrivacy: A Generative Adversarial Network for Face Anonymization
标题:DeepPrivacy:一个用于人脸匿名化的生成性对抗网络
作者: Hkon Hukkels,  Frank Lindseth
备注:Accepted to ISVC 2019
链接:https://arxiv.org/abs/1909.04538

【39】 Unrolling Ternary Neural Networks
标题:展开三元神经网络
作者: Stephen Tridgell,  Philip H.W. Leong
链接:https://arxiv.org/abs/1909.04509

【40】 ArduCode: Predictive Framework for Automation Engineering
标题:ArduCode:自动化工程预测框架
作者: Arquimedes Canedo,  Amit Pandey
链接:https://arxiv.org/abs/1909.04503

【41】 Countering Language Drift via Visual Grounding
标题:通过视觉接地对抗语言漂移
作者: Jason Lee,  Douwe Kiela
备注:Accepted to EMNLP 2019
链接:https://arxiv.org/abs/1909.04499

【42】 Natural Adversarial Sentence Generation with Gradient-based Perturbation
标题:基于梯度扰动的自然对抗性句生成
作者: Yu-Lun Hsieh,  Cho-Jui Hsieh
链接:https://arxiv.org/abs/1909.04495

【43】 Graph-based data clustering via multiscale community detection
标题:通过多尺度社区检测的基于图的数据聚类
作者: Zijing Liu,  Mauricio Barahona
链接:https://arxiv.org/abs/1909.04491

【44】 Chargrid-OCR: End-to-end trainable Optical Character Recognition through  Semantic Segmentation and Object Detection
标题:Chargrid-OCR:基于语义分割和目标检测的端到端可训练光学字符识别
作者: Christian Reisswig,  Johannes Hhne
链接:https://arxiv.org/abs/1909.04469

【45】 Automatic detection of estuarine dolphin whistles in spectrogram images
标题:光谱图像中河口海豚口哨的自动检测
作者: O. M. Serra,  L. R. Padovese
备注:10 pages; 18 figures
链接:https://arxiv.org/abs/1909.04425

【46】 Subspace clustering without knowing the number of clusters: A parameter  free approach
标题:不知道聚类数的子空间聚类:一种无参数的方法
作者: Vishnu Menon,  Sheetal Kalyani
链接:https://arxiv.org/abs/1909.04406

【47】 FDA: Feature Disruptive Attack
标题:FDA:功能中断攻击
作者: Aditya Ganeshan,  R. Venkatesh Babu
备注:Accepted in ICCV;19. Code Available at this https URL
链接:https://arxiv.org/abs/1909.04385

【48】 GBDT-MO: Gradient Boosted Decision Trees for Multiple Outputs
标题:GBDT-MO:多输出梯度增强决策树
作者: Zhendong Zhang,  Cheolkon Jung
链接:https://arxiv.org/abs/1909.04373

【49】 FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape from  Single RGB Images
标题:FreiHAND:一个用于从单个RGB图像中无标记捕获手势和形状的数据集
作者: Christian Zimmermann,  Thomas Brox
备注:Accepted to ICCV 2019, Project page: this https URL
链接:https://arxiv.org/abs/1909.04349

【50】 A Meta-Learning Framework for Generalized Zero-Shot Learning
标题:一种广义零激发学习的元学习框架
作者: Vinay Kumar Verma,  Piyush Rai
链接:https://arxiv.org/abs/1909.04344

【51】 Towards Interpretable Image Synthesis by Learning Sparsely Connected  AND-OR Networks
标题:学习稀疏连通AND-OR网络的可解释图像合成
作者: Xianglei Xing,  Ying Nian Wu
链接:https://arxiv.org/abs/1909.04324

【52】 A Bayesian Approach to Direct and Inverse Abstract Argumentation  Problems
标题:正反抽象论证问题的贝叶斯方法
作者: Hiroyuki Kido,  Beishui Liao
链接:https://arxiv.org/abs/1909.04319

【53】 Bayesian Relational Memory for Semantic Visual Navigation
标题:语义视觉导航的贝叶斯关系记忆
作者: Yi Wu,  Yuandong Tian
备注:Accepted at ICCV 2019
链接:https://arxiv.org/abs/1909.04306

【54】 Inverse Ising inference from high-temperature re-weighting of  observations
标题:高温重新加权观测值的逆Ising推断
作者: Junghyo Jo,  Vipul Periwal
链接:https://arxiv.org/abs/1909.04305

【55】 A Multistep Lyapunov Approach for Finite-Time Analysis of Biased  Stochastic Approximation
标题:有偏随机逼近有限时间分析的多步Lyapunov方法
作者: Gang Wang,  Georgios B. Giannakis
链接:https://arxiv.org/abs/1909.04299

【56】 LSTM-MSNet: Leveraging Forecasts on Sets of Related Time Series with  Multiple Seasonal Patterns
标题:LSTM-MSNet:利用具有多个季节模式的相关时间序列集合的预测
作者: Kasun Bandara,  Hansika Hewamalage
链接:https://arxiv.org/abs/1909.04293

【57】 Lifelog Patterns Analyzation using Graph Embedding based on Deep Neural  Network
标题:基于深度神经网络的图嵌入Lifelog模式分析
作者: Wonsup Shin,  Sung-Bae Cho
链接:https://arxiv.org/abs/1909.04252

【58】 PMD: A New User Distance for Recommender Systems
标题:PMD:一种新的推荐系统用户距离
作者: Yitong Meng,  Guangyong Chen
链接:https://arxiv.org/abs/1909.04239

【59】 Quantum Unsupervised and Supervised Learning on Superconducting  Processors
标题:超导处理器上的量子无监督和有监督学习
作者: Abhijat Sarma,  Ting Yu
链接:https://arxiv.org/abs/1909.04226

【60】 Machine learning accelerates parameter optimization and uncertainty  assessment of a land surface model
标题:机器学习加速陆面模型的参数优化和不确定度评估
作者: Yohei Sawada
链接:https://arxiv.org/abs/1909.04196

【61】 Recommendation System-based Upper Confidence Bound for Online  Advertising
标题:基于推荐系统的在线广告上置信限
作者: Nhan Nguyen-Thanh,  Dominique Quadri
链接:https://arxiv.org/abs/1909.04190

【62】 Signal retrieval with measurement system knowledge using variational  generative model
标题:基于变分生成模型的测量系统知识信号检索
作者: Zheyuan Zhu,  Shuo Pang
链接:https://arxiv.org/abs/1909.04188

【63】 Meta-learnt priors slow down catastrophic forgetting in neural networks
标题:元学习先验减缓神经网络中的灾难性遗忘
作者: Giacomo Spigler
链接:https://arxiv.org/abs/1909.04170

【64】 OncoNetExplainer: Explainable Predictions of Cancer Types Based on Gene  Expression Data
标题:OncoNetExplainer:基于基因表达数据的癌症类型的可解释预测
作者: Md. Rezaul Karim,  Christoph Lange
备注:In proc. of 19th IEEE International Conference on Bioinformatics and Bioengineering(IEEE BIBE 2019)
链接:https://arxiv.org/abs/1909.04169

【65】 MLOD: A multi-view 3D object detection based on robust feature fusion  method
标题:MLOD:一种基于鲁棒特征融合方法的多视角三维物体检测
作者: Jian Deng,  Krzysztof Czarnecki
备注:6 pages, 6 figures, 2019 22st International Conference on Intelligent Transportation Systems (ITSC)
链接:https://arxiv.org/abs/1909.04163

【66】 Self-Teaching Networks
标题:自学网络
作者: Liang Lu,  Yifan Gong
备注:5 pages, Interspeech 2019
链接:https://arxiv.org/abs/1909.04157

【67】 DaTscan SPECT Image Classification for Parkinson's Disease
标题:DATscan SPECT图像分类在帕金森病中的应用
作者: Justin Quan,  Jean Su
链接:https://arxiv.org/abs/1909.04142

【68】 Detection and Classification of Breast Cancer Metastates Based on U-Net
标题:基于U-NET的乳腺癌转移灶检测与分类
作者: Lin Xu,  Yu Chun Su
链接:https://arxiv.org/abs/1909.04141

【69】 Super learning for daily streamflow forecasting: Large-scale  demonstration and comparison with multiple machine learning algorithms
标题:日径流预测的超级学习:大规模演示和与多种机器学习算法的比较
作者: Hristos Tyralis,  Andreas Langousis
链接:https://arxiv.org/abs/1909.04131

【70】 DeepObfuscator: Adversarial Training Framework for Privacy-Preserving  Image Classification
标题:DeepObfuscator:保护隐私的图像分类对抗训练框架
作者: Ang Li,  Yiran Chen
链接:https://arxiv.org/abs/1909.04126

【71】 Adversarial Policy Gradient for Deep Learning Image Augmentation
标题:深度学习图像增强的对抗性策略梯度
作者: Kaiyang Cheng,  Valentina Pedoia
备注:9 pages, 2 figures, MICCAI 2019, First two authors contributed equally
链接:https://arxiv.org/abs/1909.04108

【72】 Building Calibrated Deep Models via Uncertainty Matching with Auxiliary  Interval Predictors
标题:通过与辅助区间预测器的不确定性匹配来构建校准的深度模型
作者: Jayaraman J. Thiagarajan,  Peer-Timo Bremer
链接:https://arxiv.org/abs/1909.04079

【73】 Countering the Effects of Lead Bias in News Summarization via  Multi-Stage Training and Auxiliary Losses
标题:通过多阶段训练和辅助损失对抗新闻摘要中的铅偏见影响
作者: Matt Grenander,  Annie Louis
备注:5 pages, accepted at EMNLP 2019
链接:https://arxiv.org/abs/1909.04028

机器翻译,仅供参考
分享到 :
0 人收藏
高级模式
您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

积分:43
帖子:1
精华:0
期权论坛
发布
内容

下载期权论坛手机APP