Suleman Baloch, Kishlay Jha, Alberto M. Segre, Philip M. Polgreen, Bijaya Adhikari
机构
*
Department of Electrical and Computer Engineering, University of Iowa, Iowa, USA(电气与计算机工程系,爱荷华大学,爱荷华,美国)
;
Department of Computer Science, University of Iowa, Iowa, USA(计算机科学系,爱荷华大学,爱荷华,美国)
;
Department of Internal Medicine, University of Iowa, Iowa, USA(内科学系,爱荷华大学,爱荷华,美国)
机构
*
Northeastern University, Boston, MA, USA
;
The Chinese University of Hong Kong, Hong Kong, China
;
Peking University, Beijing, China
;
Westlake University, Hangzhou, China
;
Harvard University, Cambridge, MA, USA
;
Purdue University, West Lafayette, IN, USA
;
University of Oxford, Oxford, United Kingdom
机构
*
University of California San Diego(加州大学圣地亚哥分校)
;
The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))
;
Peking University(北京大学)
;
University of California, Los Angeles(加州大学洛杉矶分校)
;
California Institute of Technology(加州理工学院)
;
ETH Zurich(苏黎世联邦理工学院)
Commentspost conference revision version at ICML; update: removed CGYRO due to bug in cases search. Will add back soon; Make the title consistent w/ pdf
CommentsPublished at the International Conference on Machine Learning (ICML) 2025. v5: earlier versions ( arXiv:2505.04608v1 (https://arxiv.org/abs/2505.04608v1) -v4 and the original ICML proceedings version) erroneously omitted an assumption (bag sufficiency) from the main theorem, which we correct here. Practical and experimental claims are unaffected. See Remark 3.5