Imprecise label learning: A unified framework for learning with various imprecise label configurations
Published in Advances in Neural Information Processing Systems 37 (NeurIPS 2024), 2024
Learning from imprecise labels is a fundamental challenge in machine learning where labels may be partial, noisy, or uncertain. This paper proposes a unified framework for learning with various imprecise label configurations, including partial labels, complementary labels, noisy labels, and weak labels. Our theoretical analysis establishes connections between different imprecise label settings, and we develop practical algorithms that can handle multiple types of label imprecision simultaneously.
Recommended citation: @inproceedings{chen2024imprecise, title={Imprecise Label Learning: A Unified Framework for Learning with Various Imprecise Label Configurations}, author={Chen, Hao and Shah, Ankit and Wang, Jiahao and Tao, Ran and Wang, Yidong and Li, Xing and Xie, Xing and Sugiyama, Masashi and Singh, Rita and Raj, Bhiksha}, booktitle={Advances in Neural Information Processing Systems 37 (NeurIPS 2024)}, year={2024} }
Download Paper