机器学习学术速递[2022.10.20]

机器学习学术速递[2022.10.20]

历史文章列表 网站 arxivdaily.com/
注:含中英文摘要速递见公众号【arXiv每日学术速递】,涵盖CS|物理|数学|经济|统计|金融|生物|电气等领域。

cs.LG 方向,今日共计124篇


Graph相关(图学习|图神经网络|图优化等)(11篇)

【1】 On Representing Mixed-Integer Linear Programs by Graph Neural Networks
标题 :用图神经网络表示混合整数线性规划
链接 arxiv.org/abs/2210.1075
作者 :Ziang Chen,Jialin Liu,Xinshang Wang,Jianfeng Lu,Wotao Yin

【2】 RSC: Accelerating Graph Neural Networks Training via Randomized Sparse Computations
标题 :RSC:通过随机稀疏计算加速图神经网络训练
链接 arxiv.org/abs/2210.1073
作者 :Zirui Liu,Shengyuan Chen,Kaixiong Zhou,Daochen Zha,Xiao Huang,Xia Hu
机构 :Rice University, United States, Hong Kong Polytechnic University

【3】 Towards Accurate Subgraph Similarity Computation via Neural Graph Pruning
标题 :基于神经网络图剪枝的精确子图相似度计算
链接 arxiv.org/abs/2210.1064
作者 :Linfeng Liu,Xu Han,Dawei Zhou,Li-Ping Liu
机构 :Meta, Boston, Department of Computer Science, Tufts University, Virginia Tech
备注 :None

【4】 DyTed: Disentangling Temporal Invariance and Fluctuations in Dynamic Graph Representation Learning
标题 :DyTed:解开动态图表示学习中的时间不变性和涨落
链接 arxiv.org/abs/2210.1059
作者 :Kaike Zhang,Qi Cao,Gaolin Fang,Bingbing Xu,Hongjian Zou,Huawei Shen,Xueqi Cheng
机构 : Data Intelligence System Research Center, Institute of Computing Technology, CAS, University of Chinese Academy of Sciences, Beijing, China, CAS Key Lab of Network Data Science and Technology, Institute of Computing Technology, CAS, Tencent Tech.

【5】 Stability of Entropic Wasserstein Barycenters and application to random geometric graphs
标题 :熵Wasserstein重心的稳定性及其在随机几何图中的应用
链接 arxiv.org/abs/2210.1053
作者 :Marc Theveneau,Nicolas Keriven

【6】 Self-supervised Heterogeneous Graph Pre-training Based on Structural Clustering
标题 :基于结构聚类的自监督异构图预训练
链接 arxiv.org/abs/2210.1046
作者 :Yaming Yang,Ziyu Guan,Zhe Wang,Wei Zhao,Cai Xu,Weigang Lu,Jianbin Huang
机构 :School of Computer Science and Technology, Xidian University
备注 :Accepted by NeurIPS 2022

【7】 EGG-GAE: scalable graph neural networks for tabular data imputation
标题 :EGG-GAE:用于表格数据填充的可伸缩图神经网络
链接 arxiv.org/abs/2210.1044
作者 :Lev Telyatnikov,Simone Scardapane
机构 :Sapienza University of Rome

【8】 Quick Graph Conversion for Robust Recommendation
标题 :用于稳健推荐的快速图表转换
链接 arxiv.org/abs/2210.1032
作者 :Zongwei Wang,Min Gao,Wentao Li
机构 :Chongqing University, China, AAII, FEIT, University of Technology, Sydney, Australia
备注 :13pages, 6 figures, 5 tables

【9】 Graph Attention Networks Unveil Determinants of Intra- and Inter-city Health Disparity
标题 :图表注意网络揭示了城市内和城市间健康差距的决定因素
链接 arxiv.org/abs/2210.1014
作者 :Chenyue Liu,Chao Fan,Ali Mostafavi
机构 : Ph.D. Student, Urban Resilience.AI Lab, Zachry Department of Civil and Environmental, College, Station, United, States;, e-mail:, Assistant Professor, Glenn Department of Civil Engineering, Clemson University, Clemson

【10】 Extending Graph Transformers with Quantum Computed Aggregation
标题 :利用量子计算聚合技术扩展图形转换器
链接 arxiv.org/abs/2210.1061
作者 :Slimane Thabet,Romain Fouilland,Loic Henriet
机构 :PASQAL SAS, Massy, France, Laboratoire d’Informatique de Paris , CNRS, Sorbonne Universit´e, Paris, France, Lo¨ıc Henriet

【11】 Graph sampling for node embedding
标题 :用于节点嵌入的图抽样方法
链接 arxiv.org/abs/2210.1052
作者 :Li-Chun Zhang

Transformer(6篇)

【1】 Transformers Learn Shortcuts to Automata
标题 :Transformer学习自动机的捷径
链接 arxiv.org/abs/2210.1074
作者 :Bingbin Liu,Jordan T. Ash,Surbhi Goel,Akshay Krishnamurthy,Cyril Zhang
机构 :Carnegie Mellon University, Microsoft Research NYC, University of Pennsylvania

【2】 RecipeMind: Guiding Ingredient Choices from Food Pairing to Recipe Completion using Cascaded Set Transformer
标题 :RecipeMind:使用级联SETTransformer指导从食物配对到食谱完成的配料选择
链接 arxiv.org/abs/2210.1062
作者 :Mogan Gim,Donghee Choi,Kana Maruyama,Jihun Choi,Hajung Kim,Donghyeon Park,Jaewoo Kang
机构 :Korea University, Seoul, Korea, Sony AI, Tokyo, Japan, Sejong University
备注 :Accepted and to be appeared in CIKM-2022

【3】 Multi-view Gait Recognition based on Siamese Vision Transformer
标题 :基于暹罗视觉变换的多视角步态识别
链接 arxiv.org/abs/2210.1042
作者 :Yanchen Yang,Lijun Yun,Ruoyu Li,Feiyan Cheng
机构 :Multi-view Gait Recognition based on Siamese Vision, Transformer, College of Information, Yunnan Normal, University, China., Yunnan Key Laboratory of Optoelectronic, Information Technology, China.
备注 :13 pages,9 figures,1 table

【4】 Museformer: Transformer with Fine- and Coarse-Grained Attention for Music Generation
标题 :MuseForm:音乐生成的细粒度和粗粒度关注的转换器
链接 arxiv.org/abs/2210.1034
作者 :Botao Yu,Peiling Lu,Rui Wang,Wei Hu,Xu Tan,Wei Ye,Shikun Zhang,Tao Qin,Tie-Yan Liu
机构 :†State Key Laboratory for Novel Software Technology, Nanjing University, China, ‡Microsoft Research Asia, §National Engineering Research Center for Software Engineering, Peking University, China
备注 :Accepted by the Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS 2022)

【5】 The Devil in Linear Transformer
标题 :线性Transformer中的魔鬼
链接 arxiv.org/abs/2210.1034
作者 :Zhen Qin,XiaoDong Han,Weixuan Sun,Dongxu Li,Lingpeng Kong,Nick Barnes,Yiran Zhong
机构 :SenseTime Research,Australian National University, OPPO Research Institute,Shanghai AI Laboratory,The University of Hong Kong
备注 :accepted to EMNLP2022

【6】 Tempo: Accelerating Transformer-Based Model Training through Memory Footprint Reduction
标题 :TEMPO:通过减少内存占用加快基于Transformer的模型训练
链接 arxiv.org/abs/2210.1024
作者 :Muralidhar Andoorveedu,Zhanda Zhu,Bojian Zheng,Gennady Pekhimenko
机构 :University of Toronto, Toronto, Canada, Shanghai Jiao Tong University, Shanghai, China, Vector Institute, Toronto, Canada
备注 :Accepted to NeurIPS 2022

GAN|对抗|攻击|生成相关(6篇)

【1】 Canary in a Coalmine: Better Membership Inference with Ensembled Adversarial Queries
标题 :煤矿中的金丝雀:利用集成的对抗性查询进行更好的成员推理
链接 arxiv.org/abs/2210.1075
作者 :Yuxin Wen,Arpit Bansal,Hamid Kazemi,Eitan Borgnia,Micah Goldblum,Jonas Geiping,Tom Goldstein
机构 :University of Maryland, University of Chicago, New York University
备注 :Code is available at this https URL

【2】 Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLP
标题 :为什么对抗性的扰动应该是不可察觉的?对抗性自然语言处理研究范式的再思考
链接 arxiv.org/abs/2210.1068
作者 :Yangyi Chen,Hongcheng Gao,Ganqu Cui,Fanchao Qi,Longtao Huang,Zhiyuan Liu,Maosong Sun
机构 :NLP Group, DCST, IAI, BNRIST, Tsinghua University, Beijing, University of Illinois Urbana-Champaign ,Chongqing University, Alibaba Group , IICTUS, Shanghai
备注 :Accepted to EMNLP 2022, main conference

【3】 Adversarial De-confounding in Individualised Treatment Effects Estimation
标题 :个体化治疗效果评估中的对抗性去混杂
链接 arxiv.org/abs/2210.1053
作者 :Vinod Kumar Chauhan,Soheila Molaei,Marzia Hoque Tania,Anshul Thakur,Tingting Zhu,David Clifton
机构 :Institute of Biomedical Engineering, University of Oxford, UK
备注 :(under review)

【4】 Few-shot Transferable Robust Representation Learning via Bilevel Attacks
标题 :基于两层攻击的少射转移稳健表示学习
链接 arxiv.org/abs/2210.1048
作者 :Minseon Kim,Hyeonjeong Ha,Sung Ju Hwang
机构 :Korea Advanced Institute of Science and Technology (KAIST),AITRICS
备注 :*Equal contribution. Author ordering determined by coin flip

【5】 Targeted Adversarial Self-Supervised Learning
标题 :定向对抗性自我监督学习
链接 arxiv.org/abs/2210.1048
作者 :Minseon Kim,Hyeonjeong Ha,Sooel Son,Sung Ju Hwang
机构 :Korea Advanced Institute of Science and Technology (KAIST),AITRICS

【6】 On the Adversarial Robustness of Mixture of Experts
标题 :关于专家混合的对抗性稳健性
链接 arxiv.org/abs/2210.1025
作者 :Joan Puigcerver,Rodolphe Jenatton,Carlos Riquelme,Pranjal Awasthi,Srinadh Bhojanapalli
机构 :Google Research
备注 :Accepted to NeurIPS 2022

半/弱/无/有监督|不确定性|主动学习(11篇)

【1】 Multi-view Tracking Using Weakly Supervised Human Motion Prediction
标题 :基于弱监督人体运动预测的多视角跟踪
链接 arxiv.org/abs/2210.1077
作者 :Martin Engilberge,Weizhe Liu,Pascal Fua
机构 :EPFL, Lausanne, Switzerland, Tencent XR Vision Labs
备注 :Accepted at WACV 2023

【2】 Weakly Supervised Learning for Analyzing Political Campaigns on Facebook
标题 :用于分析Facebook上的政治竞选的弱监督学习
链接 arxiv.org/abs/2210.1066
作者 :Tunazzina Islam,Shamik Roy,Dan Goldwasser
机构 :Department of Computer Science, Purdue University, West Lafayette, Indiana
备注 :accepted at 17th International AAAI Conference on Web and Social Media (ICWSM-2023), 12 pages

【3】 Motion-Based Weak Supervision for Video Parsing with Application to Colonoscopy
标题 :基于运动的弱监控视频解析及其在结肠镜检查中的应用
链接 arxiv.org/abs/2210.1059
作者 :Ori Kelner,Or Weinstein,Ehud Rivlin,Roman Goldenberg
机构 :Verily Life Sciences

【4】 Active Learning for Imbalanced Civil Infrastructure Data
标题 :不平衡的土木工程数据的主动学习
链接 arxiv.org/abs/2210.1058
作者 :Thomas Frick,Diego Antognini,Mattia Rigotti,Ioana Giurgiu,Benjamin Grewe,Cristiano Malossi
机构 : IBM Research, Zurich, Switzerland, Institute of Neuroinformatics, UZH and ETH Zurich, Switzerland

【5】 Propagating Variational Model Uncertainty for Bioacoustic Call Label Smoothing
标题 :生物声呼叫标签平滑的传播变分模型不确定性
链接 arxiv.org/abs/2210.1052
作者 :Georgios Rizos,Jenna Lawson,Simon Mitchell,Pranay Shah,Xin Wen,Cristina Banks-Leite,Robert Ewers,Bjoern W. Schuller
机构 :and Bj¨orn W. Schuller, GLAM – Group on Language, Audio, & Music, Department of, Computing, Imperial College London, UK., Department of Life Sciences, Imperial College London, UK., DICE – Durrell Institute of Conservation and Ecology, University of Kent, UK.

【6】 Estimating the Contamination Factor's Distribution in Unsupervised Anomaly Detection
标题 :非监督异常检测中污染因子分布的估计
链接 arxiv.org/abs/2210.1048
作者 :Lorenzo Perini,Paul Buerkner,Arto Klami
机构 :DTAI lab & Leuven.AI, Department of Computer Science, KU Leuven, B-, Leuven, Belgium, Paul-Christian Bürkner, Cluster of Excellence SimTech, University of Stuttgart, Universitätsstr. , Stuttgart, University of Helsinki, P.O.Box , Helsinki, Finland

【7】 Differentiable Self-Adaptive Learning Rate
标题 :可微自适应学习率
链接 arxiv.org/abs/2210.1029
作者 :Bozhou Chen,Hongzhi Wang,Chenmin Ba
机构 : RMSProp [6]replaces two-norm in AdaGrad with exponential moving•All authors are with the Department of Computer Science and Technology, Harbin Institute of Technology

【8】 Training set cleansing of backdoor poisoning by self-supervised representation learning
标题 :基于自监督表征学习的后门中毒训练集清理
链接 arxiv.org/abs/2210.1027
作者 :H. Wang,S. Karami,O. Dia,H. Ritter,E. Emamjomeh-Zadeh,J. Chen,Z. Xiang,D. J. Miller,G. Kesidis
机构 :Meta, Pennsylvania State University

【9】 CLUTR: Curriculum Learning via Unsupervised Task Representation Learning
标题 :CLUTR:基于无监督任务表征学习的课程学习
链接 arxiv.org/abs/2210.1024
作者 :Abdus Salam Azad,Izzeddin Gur,Aleksandra Faust,Pieter Abbeel,Ion Stoica
机构 :University of California, Berkeley, Google Research
备注 :Preprint, Currently Under Review

【10】 Uncertainty in Extreme Multi-label Classification
标题 :极端多标签分类中的不确定性
链接 arxiv.org/abs/2210.1016
作者 :Jyun-Yu Jiang,Wei-Cheng Chang,Jiong Zhong,Cho-Jui Hsieh,Hsiang-Fu Yu
机构 :Amazon Search, Palo Alto, CA , Jiong Zhang, University of California, Los Angeles, Los Angeles, CA
备注 :14 pages, 1 figure, 8 tables

【11】 A scan-specific unsupervised method for parallel MRI reconstruction via implicit neural representation
标题 :一种基于隐式神经表示的扫描特定无监督并行MRI重建方法
链接 arxiv.org/abs/2210.1043
作者 :Ruimin Feng,Qing Wu,Yuyao Zhang,Hongjiang Wei
机构 :⋆School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China;, †School of Information and Science and Technology, ShanghaiTech University, Shanghai, China;
备注 :conference

迁移|Zero/Few/One-Shot|自适应(5篇)

【1】 On the Feasibility of Cross-Task Transfer with Model-Based Reinforcement Learning
标题 :基于模型强化学习的跨任务迁移可行性研究
链接 arxiv.org/abs/2210.1076
作者 :Yifan Xu,Nicklas Hansen,Zirui Wang,Yung-Chieh Chan,Hao Su,Zhuowen Tu
机构 : University of California San Diego
备注 :Webpage with code: this https URL

【2】 Variational Model Perturbation for Source-Free Domain Adaptation
标题 :无源域自适应的变分模型摄动
链接 arxiv.org/abs/2210.1037
作者 :Mengmeng Jing,Xiantong Zhen,Jingjing Li,Cees G. M. Snoek
机构 :University of Electronic Science and Technology of China, University of Amsterdam

【3】 Adaptive Neural Network Ensemble Using Frequency Distribution
标题 :基于频率分布的自适应神经网络集成
链接 arxiv.org/abs/2210.1036
作者 :Ungki Lee,Namwoo Kang
机构 :Cho Chun Shik Graduate School of Mobility, Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea, Corresponding author

【4】 Curriculum Reinforcement Learning using Optimal Transport via Gradual Domain Adaptation
标题 :基于渐进域自适应最优传输的课程强化学习
链接 arxiv.org/abs/2210.1019
作者 :Peide Huang,Mengdi Xu,Jiacheng Zhu,Laixi Shi,Fei Fang,Ding Zhao
机构 :Carnegie Mellon University, Pittsburgh, PA
备注 :None

【5】 Deep-based quality assessment of medical images through domain adaptation
标题 :基于领域自适应的医学图像深度质量评估
链接 arxiv.org/abs/2210.1053
作者 :Marouane Tliba,Aymen Sekhri,Mohamed Amine Kerkouri,Aladine Chetouani
机构 :Laboratoire PRISME, Universit´e d’Orl´eans, Orl´eans, France, Institut National Des T´el´ecommunications et TIC, Oran, Algeria
备注 :ICIP 2022

强化学习(6篇)

【1】 When to Ask for Help: Proactive Interventions in Autonomous Reinforcement Learning
标题 :何时寻求帮助:自主强化学习中的主动干预
链接 arxiv.org/abs/2210.1076
作者 :Annie Xie,Fahim Tajwar,Archit Sharma,Chelsea Finn
机构 :Department of Computer Science, Stanford University
备注 :36th Conference on Neural Information Processing Systems (NeurIPS 2022)

【2】 Provably Safe Reinforcement Learning via Action Projection using Reachability Analysis and Polynomial Zonotopes
标题 :基于可达性分析和多项式区域规划的动作投影强化学习可证明安全
链接 arxiv.org/abs/2210.1069
作者 :Niklas Kochdumper,Hanna Krasowski,Xiao Wang,Stanley Bak,Matthias Althoff
机构 : 1Department of Computer Science, Technical University of Munich, Germany 2Department of Computer Science, Stony Brook University

【3】 Digital Human Interactive Recommendation Decision-Making Based on Reinforcement Learning
标题 :基于强化学习的数字人交互推荐决策
链接 arxiv.org/abs/2210.1063
作者 :Junwu Xiong
备注 :Preprint Under Review: NeurIPS 2022 Workshop on Human in the Loop Learning, this https URL

【4】 DIAMBRA Arena: a New Reinforcement Learning Platform for Research and Experimentation
标题 :DIAMBRA ARENA:一种新的研究和实验强化学习平台
链接 arxiv.org/abs/2210.1059
作者 :Alessandro Palmas
备注 :14 pages, 7 figures

【5】 Robust Offline Reinforcement Learning with Gradient Penalty and Constraint Relaxation
标题 :基于梯度惩罚和约束松弛的鲁棒离线强化学习
链接 arxiv.org/abs/2210.1046
作者 :Chengqian Gao,Ke Xu,Liu Liu,Deheng Ye,Peilin Zhao,Zhiqiang Xu
机构 : MBZUAI, Tencent AI Lab
备注 :9 pages, under review

【6】 A Reinforcement Learning Approach in Multi-Phase Second-Price Auction Design
标题 :强化学习在多阶段二次价格拍卖设计中的应用
链接 arxiv.org/abs/2210.1027
作者 :Rui Ai,Boxiang Lyu,Zhaoran Wang,Zhuoran Yang,Michael I. Jordan

元学习(1篇)

【1】 SML:Enhance the Network Smoothness with Skip Meta Logit for CTR Prediction
标题 :SML:通过跳过用于CTR预测的Meta Logit增强网络的平稳性
链接 arxiv.org/abs/2210.1072
作者 :Wenlong Deng,Lang Lang,Zhen Liu,Bin Liu
机构 :TikTok, China, US

蒸馏|知识提取(1篇)

【1】 Cross-Modal Fusion Distillation for Fine-Grained Sketch-Based Image Retrieval
标题 :用于基于细粒度草图的图像检索的跨模式融合提取
链接 arxiv.org/abs/2210.1048
作者 :Abhra Chaudhuri,Massimiliano Mancini,Yanbei Chen,Zeynep Akata,Anjan Dutta
机构 : University of Exeter, UK, University of Tübingen, Germany, MPI for Informatics, Germany, MPI for Intelligent Systems, Germany, University of Surrey, UK
备注 :British Machine Vision Conference (BMVC) 2022

推荐(7篇)

【1】 A Framework for Undergraduate Data Collection Strategies for Student Support Recommendation Systems in Higher Education
标题 :高等教育学生支持推荐系统的本科生数据收集策略框架
链接 arxiv.org/abs/2210.1065
作者 :Herkulaas MvE Combrink,Vukosi Marivate,Benjamin Rosman
备注 :14 pages, 4 figures, Proceedings of the 2020 SACAIR Conference

【2】 Multi-Modal Recommendation System with Auxiliary Information
标题 :具有辅助信息的多通道推荐系统
链接 arxiv.org/abs/2210.1065
作者 :Mufhumudzi Muthivhi,Terence L. van Zyl,Hairong Wang
机构 : University of the Witwatersrand, Johannesburg GT , South Africa, University of Johannesburg, Johannesburg GT, South Africa
备注 :15 pages, 3 figures, 3 tables, to be published in the SACAIR CCIS Springer proceedings volume

【3】 Using Interventions to Improve Out-of-Distribution Generalization of Text-Matching Recommendation Systems
标题 :利用干预改进文本匹配推荐系统的离散型泛化
链接 arxiv.org/abs/2210.1063
作者 :Parikshit Bansal,Yashoteja Prabhu,Emre Kiciman,Amit Sharma
机构 :Microsoft Research, Emre Kıcıman

【4】 Simulated Contextual Bandits for Personalization Tasks from Recommendation Datasets
标题 :来自推荐数据集的个性化任务的模拟上下文Bitits
链接 arxiv.org/abs/2210.1063
作者 :Anton Dereventsov,Anton Bibin
机构 :Lirio AI Research, Lirio, LLC, Knoxville, TN, USA, Skoltech Agro, Moscow, Russia

【5】 Multi-Objective Recommender Systems: Survey and Challenges
标题 :多目标推荐系统:综述与挑战
链接 arxiv.org/abs/2210.1030
作者 :Dietmar Jannach
机构 :University of Klagenfurt, Austria
备注 :In: Proceedings of the 2nd Workshop on Multi-Objective Recommender Systems (MORS) held in conjunction with the 16th ACM Conference on Recommender Systems (RecSys), 2022, Seattle, USA

【6】 Causal Structure Learning with Recommendation System
标题 :基于推荐系统的因果结构学习
链接 arxiv.org/abs/2210.1025
作者 :Shuyuan Xu,Da Xu,Evren Korpeoglu,Sushant Kumar,Stephen Guo,Kannan Achan,Yongfeng Zhang
机构 :†Rutgers University, ‡Walmart Labs, §Indeed

【7】 Auditing YouTube's Recommendation Algorithm for Misinformation Filter Bubbles
标题 :审计YouTube的虚假信息过滤气泡推荐算法
链接 arxiv.org/abs/2210.1008
作者 :Ivan Srba,Robert Moro,Matus Tomlein,Branislav Pecher,Jakub Simko,Elena Stefancova,Michal Kompan,Andrea Hrckova,Juraj Podrouzek,Adrian Gavornik,Maria Bielikova
机构 : Kempelen Institute of Intelligent Technologies, Brno University of Technology
备注 :Just accepted to ACM Transactions on Recommender Systems (ACM TORS). arXiv admin note: substantial text overlap with arXiv:2203.13769

聚类(1篇)

【1】 Towards Practical Explainability with Cluster Descriptors
标题 :使用簇描述符实现实用的可解释性
链接 arxiv.org/abs/2210.1066
作者 :Xiaoyuan Liu,Ilya Tyagin,Hayato Ushijima-Mwesigwa,Indradeep Ghosh,Ilya Safro
机构 :Fujitsu Research of America, Inc., Sunnyvale, CA, USA, University of Delaware, Newark, DE, USA

超分辨率|去噪|去模糊|去雾(1篇)

【1】 Explainable bilevel optimization: an application to the Helsinki deblur challenge
标题 :可解释的双层优化:在赫尔辛基反模糊挑战中的应用
链接 arxiv.org/abs/2210.1005
作者 :Silvia Bonettini,Giorgia Franchini,Danilo Pezzi,Marco Prato
机构 :Dipartimento di Scienze Fisiche, Informatiche e Matematiche, Universita di Modena e Reggio Emilia, Via Campi ,b - , Modena (MO), Italy, All the authors are members of the INdAM research group GNCS

联邦学习|隐私保护|加密(1篇)

【1】 TEFL: Turbo Explainable Federated Learning for 6G Trustworthy Zero-Touch Network Slicing
标题 :TEFL:6G可信零接触网络切片的Turbo可解释联邦学习
链接 arxiv.org/abs/2210.1014
作者 :Swastika Roy,Hatim Chergui,Christos Verikoukis
机构 : Verikoukisis with University of Patras

推理|分析|理解|解释(6篇)

【1】 CLEAR: Causal Explanations from Attention in Neural Recommenders
标题 :明确:神经推荐器中注意力的因果解释
链接 arxiv.org/abs/2210.1062
作者 :Shami Nisimov,Raanan Y. Rohekar,Yaniv Gurwicz,Guy Koren,Gal Novik
备注 :Causality, Counterfactuals and Sequential Decision-Making for Recommender Systems (CONSEQUENCES) workshop at RecSys 2022, Seattle, WA, USA

【2】 Explainable Slot Type Attentions to Improve Joint Intent Detection and Slot Filling
标题 :解释槽型注意改进联合意图检测和槽位填充
链接 arxiv.org/abs/2210.1022
作者 :Kalpa Gunaratna,Vijay Srinivasan,Akhila Yerukola,Hongxia Jin
机构 :†Samsung Research America, Mountain View CA, USA, ‡Carnegie Mellon University, Pittsburgh PA, USA
备注 :EMNLP 2022

【3】 Granger causal inference on DAGs identifies genomic loci regulating transcription
标题 :DAGs上的Granger因果推断确定了调控转录的基因组位点
链接 arxiv.org/abs/2210.1016
作者 :Rohit Singh,Alexander P. Wu,Bonnie Berger
机构 : Computer Science and Artificial Intelligence Laboratory, MIT, Cambridge, MA , USA, Department of Mathematics, MIT, Cambridge, MA , USA, † Authors contributed equally to this work
备注 :Presented at the International Conference on Learning Representations 2022

【4】 Anytime-valid off-policy inference for contextual bandits
标题 :针对上下文盗贼的随时有效的非策略推理
链接 arxiv.org/abs/2210.1076
作者 :Ian Waudby-Smith,Lili Wu,Aaditya Ramdas,Nikos Karampatziakis,Paul Mineiro
机构 :Carnegie Mellon University, Microsoft

【5】 Rethinking Sharpness-Aware Minimization as Variational Inference
标题 :将锐度感知最小化作为变分推理的再思考
链接 arxiv.org/abs/2210.1045
作者 :Szilvia Ujváry,Zsigmond Telek,Anna Kerekes,Anna Mészáros,Ferenc Huszár
机构 :Anna Mesz´aros∗, Ferenc Husz´ar

【6】 Inference in conditioned dynamics through causality restoration
标题 :基于因果关系恢复的条件动力学推理
链接 arxiv.org/abs/2210.1017
作者 :Alfredo Braunstein,Giovanni Catania,Luca Dall'Asta,Matteo Mariani,Anna Paola Muntoni
机构 :DISAT, Politecnico di Torino, Corso Duca Degli Abruzzi , Torino, Collegio Carlo Alberto, P.za Arbarello , Torino, Italy, Italian Institute for Genomic Medicine, IRCCS Candiolo, SP-, I-, Candiolo (TO), Italy
备注 :22 pages, 7 figures

检测相关(4篇)

【1】 Anomaly Detection Requires Better Representations
标题 :异常检测需要更好的表示
链接 arxiv.org/abs/2210.1077
作者 :Tal Reiss,Niv Cohen,Eliahu Horwitz,Ron Abutbul,Yedid Hoshen
机构 :School of Computer Science and Engineering, The Hebrew University of Jerusalem, Israel
备注 :Accepted to ECCV SSLWIN Workshop (2022)

【2】 Two-level Data Augmentation for Calibrated Multi-view Detection
标题 :用于定标多视点检测的两级数据增强
链接 arxiv.org/abs/2210.1075
作者 :Martin Engilberge,Haixin Shi,Zhiye Wang,Pascal Fua
机构 :EPFL, Lausanne, Switzerland
备注 :Accepted at WACV 2023

【3】 Non-iterative optimization of pseudo-labeling thresholds for training object detection models from multiple datasets
标题 :多数据集中训练目标检测模型伪标签阈值的非迭代优化
链接 arxiv.org/abs/2210.1022
作者 :Yuki Tanaka,Shuhei M. Yoshida,Makoto Terao
机构 :Visual Intelligence Research Laboratories, NEC Corporation, Kawasaki, Kanagawa, Japan
备注 :ICIP2022

【4】 Detecting and analyzing missing citations to published scientific entities
标题 :检测和分析对已发表科学实体的遗漏引文
链接 arxiv.org/abs/2210.1007
作者 :Jialiang Lin,Yao Yu,Jiaxin Song,Xiaodong Shi
机构 :School of Informatics, Xiamen University, Xiamen, China.
备注 :None

分类|识别(3篇)

【1】 AUC-based Selective Classification
标题 :基于AUC的选择性分类
链接 arxiv.org/abs/2210.1070
作者 :Andrea Pugnana,Salvatore Ruggieri
机构 :Scuola Normale Superiore, University of Pisa

【2】 Using deep convolutional neural networks to classify poisonous and edible mushrooms found in China
标题 :利用深层卷积神经网络对中国有毒食用菌进行分类
链接 arxiv.org/abs/2210.1035
作者 :Baiming Zhang,Ying Zhao,Zhixiang Li
机构 :Northeast Yucai Foreign Language School, Shenyang, Liaoning, China, Northwestern University, Evanston, IL, USA, Department of Biomedical Engineering, Shenyang Pharmaceutical University
备注 :5 pages, 2 figures

【3】 Margin Optimal Classification Trees
标题 :边际最优分类树
链接 arxiv.org/abs/2210.1056
作者 :Federico D'Onofrio,Giorgio Grani,Marta Monaci,Laura Palagi
机构 :Department of Computer, Control and Management Engineering ”Antonio Ruberti” (DIAG), Department of Statistical Sciences, Sapienza University of Rome, Rome, Italy

表征(1篇)

【1】 Deep Multi-Representation Model for Click-Through Rate Prediction
标题 :点击率预测的深度多表示模型
链接 arxiv.org/abs/2210.1066
作者 :Shereen Elsayed,Lars Schmidt-Thieme
机构 :Information Systems and Machine Learning Lab, Hildesheim, Germany

3D|3D重建等相关(1篇)

【1】 Estimating the coverage in 3d reconstructions of the colon from colonoscopy videos
标题 :从结肠镜视频估计结肠3D重建的覆盖率
链接 arxiv.org/abs/2210.1045
作者 :Emmanuelle Muhlethaler,Erez Posner,Moshe Bouhnik
机构 :Intuitive Surgical, Inc., Kifer Road, Sunnyvale, CA
备注 :Accepted at Imaging Systems for GI Endoscopy workshop, the 25th International Conference on Medical Image Computing and Computer Assisted Intervention - MICCAI 2022 ISGIE

优化|敛散性(2篇)

【1】 Bayesian Optimization over Discrete and Mixed Spaces via Probabilistic Reparameterization
标题 :基于概率重参数的离散和混合空间贝叶斯优化
链接 arxiv.org/abs/2210.1019
作者 :Samuel Daulton,Xingchen Wan,David Eriksson,Maximilian Balandat,Michael A. Osborne,Eytan Bakshy
机构 :University of Oxford, Meta
备注 :To appear in Advances in Neural Information Processing Systems 35, 2022. Code available at: this https URL

【2】 Deep neural network expressivity for optimal stopping problems
标题 :最优停止理论问题的深度神经网络表达能力
链接 arxiv.org/abs/2210.1044
作者 :Lukas Gonon

预测|估计(6篇)

【1】 Autoregressive Generative Modeling with Noise Conditional Maximum Likelihood Estimation
标题 :带噪声条件极大似然估计的自回归生成模型
链接 arxiv.org/abs/2210.1071
作者 :Henry Li,Yuval Kluger
机构 :Yale University
备注 :18 pages, 10 figures, 2 tables

【2】 A Segment-Wise Gaussian Process-Based Ground Segmentation With Local Smoothness Estimation
标题 :基于分段高斯过程的局部光滑度估计的地面分割
链接 arxiv.org/abs/2210.1051
作者 :Pouria Mehrabi,Hamid D. Taghirad
机构 :Advance Robotics & Automated Systems (ARAS), K. N. Toosi University of Technology, Tehran, Iran.

【3】 Group Fairness in Prediction-Based Decision Making: From Moral Assessment to Implementation
标题 :预测型决策中的群体公平:从道德评估到实施
链接 arxiv.org/abs/2210.1045
作者 :Joachim Baumann,Christoph Heitz
机构 :University of Zurich, Zurich University of Applied Sciences, Zurich, Switzerland, Winterthur, Switzerland
备注 :None

【4】 Comparing Spectroscopy Measurements in the Prediction of in Vitro Dissolution Profile using Artificial Neural Networks
标题 :用人工神经网络预测体外溶出度的光谱测量方法比较
链接 arxiv.org/abs/2210.1029
作者 :Mohamed Azouz Mrad,Kristóf Csorba,Dorián László Galata,Zsombor Kristóf Nagy,Brigitta Nagy
机构 :Department of Automation and Applied Informatics, Budapest University of Technology and Economics, Budapest, Hungary
备注 :11 pagee, 5 figures, 3rd International Conference on Data Science and Machine Learning (DSML 2022)

【5】 Predicting Oxide Glass Properties with Low Complexity Neural Network and Physical and Chemical Descriptors
标题 :基于低复杂度神经网络和物理化学参数的氧化物玻璃性能预测
链接 arxiv.org/abs/2210.1050
作者 :Suresh Bishnoi,Skyler Badge,Jayadeva,N. M. Anoop Krishnan
机构 :School of Interdisciplinary Research, Indian Institute of Technology Delhi, Hauz Khas, New, Department of Electrical Engineering, Indian Institute of Technology Delhi, Hauz Khas, New Delhi , India
备注 :15 pages, 3 figures

【6】 Nonparametric Quantile Regression: Non-Crossing Constraints and Conformal Prediction
标题 :非参数分位数回归:非交叉约束与保角预测
链接 arxiv.org/abs/2210.1016
作者 :Wenlu Tang,Guohao Shen,Yuanyuan Lin,Jian Huang
备注 :8 figures, 3 tables

其他神经网络|深度学习|模型|建模(24篇)

【1】 "Why did the Model Fail?": Attributing Model Performance Changes to Distribution Shifts
链接 arxiv.org/abs/2210.1076
作者 :Haoran Zhang,Harvineet Singh,Marzyeh Ghassemi,Shalmali Joshi
机构 :MIT, New York University, Harvard University

【2】 Scaling Laws for Reward Model Overoptimization
标题 :报酬模型过优化的标度律
链接 arxiv.org/abs/2210.1076
作者 :Leo Gao,John Schulman,Jacob Hilton
机构 :OpenAI

【3】 Robustness of Demonstration-based Learning Under Limited Data Scenario
标题 :有限数据场景下基于演示学习的稳健性
链接 arxiv.org/abs/2210.1069
作者 :Hongxin Zhang,Yanzhe Zhang,Ruiyi Zhang,Diyi Yang
机构 :Shanghai Jiao Tong University,Georgia Institute of Technology, Adobe Research,Stanford University
备注 :14 pages, EMNLP 2022 Main Conference

【4】 Attaining Class-level Forgetting in Pretrained Model using Few Samples
标题 :利用小样本实现预训练模型中的类级遗忘
链接 arxiv.org/abs/2210.1067
作者 :Pravendra Singh,Pratik Mazumder,Mohammed Asad Karim
机构 :IIT Roorkee, India, IIT Kanpur, India, Carnegie Mellon University, USA
备注 :Accepted in ECCV 2022

【5】 Fantômas: Evaluating Reversibility of Face Anonymizations Using a General Deep Learning Attacker
标题 :使用一般深度学习攻击者评估人脸匿名化的可逆性
链接 arxiv.org/abs/2210.1065
作者 :Julian Todt,Simon Hanisch,Thorsten Strufe
机构 :KASTEL Security Research Labs, Karlsruhe Institute of Technology, Centre for Tactile Internet with Human-in-the-Loop (CeTI), Technical University Dresden

【6】 Robust Regression with Highly Corrupted Data via Physics Informed Neural Networks
标题 :基于物理信息神经网络的高污染数据稳健回归
链接 arxiv.org/abs/2210.1064
作者 :Wei Peng,Wen Yao,Weien Zhou,Xiaoya Zhang,Weijie Yao
机构 :Defense Innovation Institute, Chinese Academy of Military Science, Beijing, China, College of Aerospace Science and Engineering, National University of Defense Technology, Changsha, China

【7】 Irregularly-Sampled Time Series Modeling with Spline Networks
标题 :基于样条网的非规则抽样时间序列建模
链接 arxiv.org/abs/2210.1063
作者 :Marin Biloš,Emanuel Ramneantu,Stephan Günnemann
机构 :Technical University of Munich, Germany
备注 :ICML Continuous Time Methods for Machine Learning Workshop (2022)

【8】 DALLE-2 is Seeing Double: Flaws in Word-to-Concept Mapping in Text2Image Models
标题 :Dalle-2看到了双重:Text2Image模型中单词到概念映射的缺陷
链接 arxiv.org/abs/2210.1060
作者 :Royi Rassin,Shauli Ravfogel,Yoav Goldberg
机构 :Bar-Ilan University, Allen Institute for Artificial Intelligence
备注 :5 pages, BlackboxNLP @ EMNLP 2022

【9】 Geometric Deep Learning for the Assessment of Thrombosis Risk in the Left Atrial Appendage
标题 :几何深度学习在左心耳血栓风险评估中的应用
链接 arxiv.org/abs/2210.1056
作者 :Xabier Morales,Jordi Mill,Guillem Simeon,Kristine A. Juhl,Ole De Backer,Rasmus R. Paulsen,Oscar Camara
机构 : Physense, BCN Medtech, Department of Information and Communications, Technologies, Universitat Pompeu Fabra, Barcelona, Spain, DTU Compute, Technical University of Denmark, Kongens Lyngby, Denmark., Heart Center, Rigshospitalet, Copenhagen, Denmark.

【10】 Hierarchical Multi-Interest Co-Network For Coarse-Grained Ranking
标题 :一种用于粗粒度排序的层次化多兴趣联合网络
链接 arxiv.org/abs/2210.1054
作者 :Xu Yuan,Chen Xu,Qiwei Chen,Tao Zhuang,Hongjie Chen,Chao Li,Junfeng Ge
机构 :Alibaba Group, Beijing, China

【11】 Language Model Decomposition: Quantifying the Dependency and Correlation of Language Models
标题 :语言模型分解:量化语言模型的依赖性和相关性
链接 arxiv.org/abs/2210.1028
作者 :Hao Zhang
机构 :Supportiv Inc, Berkeley, CA
备注 :accepted by EMNLP 2022

【12】 A new activation for neural networks and its approximation
标题 :一种新的神经网络激活方法及其逼近
链接 arxiv.org/abs/2210.1026
作者 :Jianfei Li,Han Feng,Ding-Xuan Zhou

【13】 Discovering Limitations of Image Quality Assessments with Noised Deep Learning Image Sets
标题 :利用带噪声的深度学习图像集发现图像质量评估的局限性
链接 arxiv.org/abs/2210.1024
作者 :Wei Dai,Daniel Berleant
机构 :Department of Computer Science, Purdue University Northwest, Hammond, Indiana, USA, Department of Information Science, University of Arkansas at Little Rock, Little Rock, Arkansas, USA
备注 :10 pages, 11 figures, 10 tables

【14】 Exclusive Supermask Subnetwork Training for Continual Learning
标题 :专为持续学习而设计的超级屏蔽子网络训练
链接 arxiv.org/abs/2210.1020
作者 :Prateek Yadav,Mohit Bansal
机构 :Department of Computer Science, UNC Chapel Hill
备注 :17 pages, 7 figures

【15】 Rethinking Prototypical Contrastive Learning through Alignment, Uniformity and Correlation
标题 :从对齐、一致和关联中反思原型对比学习
链接 arxiv.org/abs/2210.1019
作者 :Shentong Mo,Zhun Sun,Chao Li
机构 :Rethinking Prototypical Contrastive, Learning, through, Alignment, Uniformity and Correlation, Carnegie Mellon University, Pittsburgh, PA , United States, Tohoku University, Sendai, Miyagi, Japan, Center for Advanced Intelligence, Project (AIP), RIKEN, Tokyo, Japan
备注 :BMVC 2022

【16】 High-Dimensional Performance Modeling via Tensor Completion
标题 :基于张量补全的高维性能建模
链接 arxiv.org/abs/2210.1018
作者 :Edward Hutter,Edgar Solomonik
机构 :Department of Computer Science, University of Illinois at Urbana-Champaign

【17】 Output Feedback Tube MPC-Guided Data Augmentation for Robust, Efficient Sensorimotor Policy Learning
标题 :输出反馈管MPC引导的数据增强,用于稳健、高效的感觉运动策略学习
链接 arxiv.org/abs/2210.1012
作者 :Andrea Tagliabue,Jonathan P. How
机构 : which can cause deviationsAll the authors are with the MIT Department of Aeronautics andAstronautics
备注 :Accepted to IROS 22

【18】 Proximal Learning With Opponent-Learning Awareness
标题 :近距离学习与对手--学习意识
链接 arxiv.org/abs/2210.1012
作者 :Stephen Zhao,Chris Lu,Roger Baker Grosse,Jakob Nicolaus Foerster
机构 :University of Toronto and Vector Institute, FLAIR, University of Oxford, Roger Grosse, Jakob Foerster
备注 :24 pages (10 pages main paper), 5 figures, to be published in 36th Conference on Neural Information Processing Systems (NeurIPS 2022)

【19】 Optimisation & Generalisation in Networks of Neurons
标题 :神经元网络中的优化与泛化
链接 arxiv.org/abs/2210.1010
作者 :Jeremy Bernstein
机构 :In Partial Fulfillment of the Requirements for, the Degree of Doctor of Philosophy, California Institute of Technology, Pasadena, California, arXiv:,.,v, [cs.NE] , Oct
备注 :PhD thesis

【20】 AnalogVNN: A fully modular framework for modeling and optimizing photonic neural networks
标题 :AnalogVNN:一种全模块化的光子神经网络建模与优化框架
链接 arxiv.org/abs/2210.1004
作者 :Vivswan Shah,Nathan Youngblood
机构 :Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA

【21】 A kernel Stein test of goodness of fit for sequential models
标题 :序贯模型拟合优度的核Stein检验
链接 arxiv.org/abs/2210.1074
作者 :Jerome Baum,Heishiro Kanagawa,Arthur Gretton
机构 :University College London, Gatsby Computational Neuroscience Unit
备注 :13 pages

【22】 The phase unwrapping of under-sampled interferograms using radial basis function neural networks
标题 :基于径向基函数神经网络的欠采样干涉图相位展开
链接 arxiv.org/abs/2210.1054
作者 :Pierre-Alexandre Gourdain,Aidan Bachmann
机构 :University of Rochester

【23】 Spectroscopic data de-noising via training-set-free deep learning method
标题 :基于无训练集深度学习的光谱数据去噪方法
链接 arxiv.org/abs/2210.1049
作者 :Dongchen Huang,Junde Liu,Tian Qian,Yi-feng Yang
机构 :Beijing National Laboratory for Condensed Matter Physics and Institute of Physics, University of Chinese Academy of Sciences, Beijing , China, Songshan Lake Materials Laboratory, Dongguan, Guangdong , China

【24】 Machine Learning for a Sustainable Energy Future
标题 :面向可持续能源未来的机器学习
链接 arxiv.org/abs/2210.1039
作者 :Zhenpeng Yao,Yanwei Lum,Andrew Johnston,Luis Martin Mejia-Mendoza,Xin Zhou,Yonggang Wen,Alan Aspuru-Guzik,Edward H. Sargent,Zhi Wei Seh
机构 :Center of Hydrogen Science, Shanghai Jiao, Tong University, Dongchuan Road, Chemical Physics Theory Group, Department of Chemistry and Department of, Computer Science, University of Toronto, Toronto, Ontario M,S ,H, Canada, Institute of Materials Research and

其他(20篇)

【1】 OpenEarthMap: A Benchmark Dataset for Global High-Resolution Land Cover Mapping
标题 :OpenEarthMap:全球高分辨率土地覆盖制图的基准数据集
链接 arxiv.org/abs/2210.1073
作者 :Junshi Xia,Naoto Yokoya,Bruno Adriano,Clifford Broni-Bediako
机构 :jp 2The University of Tokyo
备注 :Accepted by WACV 2023

【2】 Binary Orthogonal Non-negative Matrix Factorization
标题 :二元正交非负矩阵分解
链接 arxiv.org/abs/2210.1066
作者 :S. Fathi Hafshejani,D. Gaur,S. Hossain,R. Benkoczi
机构 :Department of Math and Computer Science, University of Lethbridge, Lethbridge, AB, Canada, sajad.fathihafshejan, daya.gaur

【3】 Review of the state of the art in autonomous artificial intelligence
标题 :自主人工智能研究现状综述
链接 arxiv.org/abs/2210.1065
作者 :Petar Radanliev,David De Roure
机构 :Department of Engineering Sciences, University of Oxford, Oxford OX,QG, England, UK
备注 :AI Ethics (2022)

【4】 Controlling Travel Path of Original Cobra
标题 :控制原始眼镜蛇的行进路径
链接 arxiv.org/abs/2210.1065
作者 :Mriganka Basu RoyChowdhury,Arabin K Dey
机构 :a Department of Mathematics, University of California, Berkley; b Department of, Mathematics, IIT Guwahati;, ARTICLE HISTORY
备注 :9 pages; 2 figures

【5】 POGD: Gradient Descent with New Stochastic Rules
标题 :POGD:新的随机规则下的梯度下降
链接 arxiv.org/abs/2210.1065
作者 :Feihu Han,Sida Xing,Sui Yang Khoo
机构 :Deakin University

【6】 HyperMiner: Topic Taxonomy Mining with Hyperbolic Embedding
标题 :HyperMiner:基于双曲嵌入的主题分类挖掘
链接 arxiv.org/abs/2210.1062
作者 :Yishi Xu,Dongsheng Wang,Bo Chen,Ruiying Lu,Zhibin Duan,Mingyuan Zhou
机构 :National Laboratory of Radar Signal Processing, Xidian University, Xi’an, China, McCombs School of Business, The University of Texas at Austin, USA

【7】 Active Image Indexing
标题 :活动图像索引
链接 arxiv.org/abs/2210.1062
作者 :Pierre Fernandez,Matthijs Douze,Hervé Jégou,Teddy Furon
机构 :Meta AI, FAIR, Univ. Rennes, Inria, CNRS, IRISA

【8】 Enhanced vectors for top-k document retrieval in Question Answering
标题 :用于问答中前k个文档检索的增强型向量
链接 arxiv.org/abs/2210.1058
作者 :Mohammed Hammad
备注 :Year-2019

【9】 The Future of Consumer Edge-AI Computing
标题 :消费者边缘的未来--AI计算
链接 arxiv.org/abs/2210.1051
作者 :Stefanos Laskaridis,Stylianos I. Venieris,Alexandros Kouris,Rui Li,Nicholas D. Lane
机构 :†Samsung AI Center, Cambridge, UK, ‡University of Cambridge
备注 :Under review

【10】 Attribution and Obfuscation of Neural Text Authorship: A Data Mining Perspective
标题 :基于数据挖掘的神经文本作者归因与混淆
链接 arxiv.org/abs/2210.1048
作者 :Adaku Uchendu,Thai Le,Dongwon Lee
机构 :The Pennsylvania State University, University Park, PA, USA, The University of Mississippi, Oxford, MS, USA

【11】 On the Power of Pre-training for Generalization in RL: Provable Benefits and Hardness
标题 :RL泛化前训练的威力:可证明的益处和难度
链接 arxiv.org/abs/2210.1046
作者 :Haotian Ye,Xiaoyu Chen,Liwei Wang,Simon S. Du

【12】 Gaussian-Bernoulli RBMs Without Tears
标题 :无撕裂的Gauss-Bernoulli RBM
链接 arxiv.org/abs/2210.1031
作者 :Renjie Liao,Simon Kornblith,Mengye Ren,David J. Fleet,Geoffrey Hinton
机构 :University of British Columbia, Google Research, Brain Team, New York University, University of Toronto, Vector Institute

【13】 Towards Explaining Distribution Shifts
标题 :向分配转移的解释迈进
链接 arxiv.org/abs/2210.1027
作者 :Sean Kulinski,David I. Inouye
机构 :Purdue University

【14】 A Catch-22 of Reservoir Computing
标题 :水库计算的第22条军规
链接 arxiv.org/abs/2210.1021
作者 :Yuanzhao Zhang,Sean P. Cornelius
机构 :Santa Fe Institute, Hyde Park Road, Santa Fe, NM , USA, Department of Physics, Toronto Metropolitan University, Toronto, ON, M,B ,K, Canada
备注 :Comments welcome. Our code can be found at this https URL

【15】 Optimizing Hierarchical Image VAEs for Sample Quality
标题 :基于样本质量的层次化图像VAE优化
链接 arxiv.org/abs/2210.1020
作者 :Eric Luhman,Troy Luhman
备注 :21 pages, 12 figures

【16】 Intra-Source Style Augmentation for Improved Domain Generalization
标题 :用于改进领域泛化的源代码内样式增强
链接 arxiv.org/abs/2210.1017
作者 :Yumeng Li,Dan Zhang,Margret Keuper,Anna Khoreva
机构 :Bosch Center for AI, University of Siegen, MPI for Informatics, University of T¨ubingen
备注 :Accepted at WACV 2023

【17】 Transferable Unlearnable Examples
标题 :可迁移的不可学实例
链接 arxiv.org/abs/2210.1011
作者 :Jie Ren,Han Xu,Yuxuan Wan,Xingjun Ma,Lichao Sun,Jiliang Tang
机构 :Michigan State University, Fudan University, Lehigh University

【18】 Convexity Certificates from Hessians
标题 :来自黑森州的凸性证书
链接 arxiv.org/abs/2210.1043
作者 :Julien Klaus,Niklas Merk,Konstantin Wiedom,Sören Laue,Joachim Giesen
机构 :Friedrich Schiller University Jena, Technical University of Kaiserslautern, Data Assessment Solutions GmbH
备注 :Accepted for publication at NeurIPS 2022

【19】 Fast Approximation of the Generalized Sliced-Wasserstein Distance
标题 :广义分片-瓦瑟斯坦距离的快速逼近
链接 arxiv.org/abs/2210.1026
作者 :Dung Le,Huy Nguyen,Khai Nguyen,Trang Nguyen,Nhat Ho
机构 :École Polytechnique⋄; University of Texas, Austin†;, Hanoi University of Science and Technology§
备注 :22 pages, 2 figures. Dung Le, Huy Nguyen and Khai Nguyen contributed equally to this work

【20】 Combination of Raman spectroscopy and chemometrics: A review of recent studies published in the Spectrochimica Acta, Part A: Molecular and Biomolecular Spectroscopy Journal
标题 :拉曼光谱和化学计量学的结合:《光谱学报》A部分:分子和生物分子光谱杂志最近发表的研究综述
链接 arxiv.org/abs/2210.1005
作者 :Yulia Khristoforova,Lyudmila Bratchenko,Ivan Bratchenko
机构 : Bratchenko Samara National Research University
备注 :23 pages, 1 figure

机器翻译由腾讯交互翻译提供,仅供参考


历史文章列表 网站 arxivdaily.com/
注:含中英文摘要速递见公众号【arXiv每日学术速递】,涵盖CS|物理|数学|经济|统计|金融|生物|电气等领域。
发布于 2022-10-20 10:45

文章被以下专栏收录