Learning with Complementary Labels Revisited: A Consistent Approach via Negative-Unlabeled Learning

Wei Wang, Takashi Ishida, Yu-Jie Zhang, Gang Niu, Masashi Sugiyama

Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: Complementary-label learning, weakly supervised learning, positive-unlabeled learning

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: A consistent complementary-label learning approach without relying on the uniform distribution assumption or an ordinary-label training set to estimate the transition matrix.

Abstract: Complementary-label learning is a weakly supervised learning problem in which each training example is associated with one or multiple complementary labels indicating the classes to which it does not belong. Existing consistent approaches have relied on the uniform distribution assumption to model the generation of complementary labels, or on an ordinary-label training set to estimate the transition matrix. However, both conditions may not be satisfied in real-world scenarios. In this paper, we propose a novel complementary-label learning approach that does not rely on these conditions. We find that complementary-label learning can be expressed as a set of negative-unlabeled binary classification problems when using the one-versus-rest strategy. This observation allows us to propose a risk-consistent approach with theoretical guarantees. Furthermore, we introduce a risk correction approach to address overfitting problems when using complex models. We also prove the statistical consistency and convergence rate of the corrected risk estimator. Extensive experimental results on both synthetic and real-world benchmark datasets validate the superiority of our proposed approach over state-of-the-art methods.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 2453