ICML-RATT Leveraging Unlabeled Data to Guarantee Generalization

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

RATT: Leveraging Unlabeled Data to Guarantee Generalization

         Saurabh Garg 1 Sivaraman Balakrishnan 1 2 J. Zico Kolter 3 Zachary C. Lipton 1

            Abstract                ity of the hypothesis class. After fitting the model on the
   To assess generalization, machine learning scien-     available data, one can plug in the empirical risk to obtain
   tists typically either (i) bound the generalization    a guarantee on the true risk. The second approach, favored
   gap and then (af ...

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Guarantee General label Aging ATION