Confidence Scores Make Instance-dependent Label-noise Learning Possible

Antonin Berthon, Bo HAN, Gang Niu, Tongliang Liu, Masashi Sugiyama

Research output: Chapter in book/report/conference proceedingConference contributionpeer-review

Abstract

In learning with noisy labels, for every instance, its label can randomly walk to other classes following a transition distribution which is named a noise model. Well-studied noise models are all instance-independent, namely, the transition depends only on the original label but not the instance itself, and thus they are less practical in the wild. Fortunately, methods based on instance-dependent noise have been studied, but most of them have to rely on strong assumptions on the noise models. To alleviate this issue, we introduce confidence-scored instance-dependent noise (CSIDN), where each instance-label pair is equipped with a confidence score. We find that with the help of confidence scores, the transition distribution of each instance can be approximately estimated. Similarly to the powerful forward correction for instance-independent noise, we propose a novel instance-level forward correction for CSIDN. We demonstrate the utility and effectiveness of our method through multiple experiments on datasets with synthetic label noise and real-world unknown noise.
Original languageEnglish
Title of host publicationProceedings of the 38th International Conference on Machine Learning
PublisherML Research Press
Pages825-836
Number of pages12
Publication statusPublished - 18 Jul 2021
Event38th International Conference on Machine Learning, ICML 2021 - Virtual
Duration: 18 Jul 202124 Jul 2021
https://icml.cc/Conferences/2021

Publication series

NameProceedings of Machine Learning Research
Volume138
ISSN (Print)2640-3498

Conference

Conference38th International Conference on Machine Learning, ICML 2021
Period18/07/2124/07/21
Internet address

Fingerprint

Dive into the research topics of 'Confidence Scores Make Instance-dependent Label-noise Learning Possible'. Together they form a unique fingerprint.

Cite this