CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learning

Chen Wei, Kihyuk Sohn, Clayton Mellina, Alan Yuille, Fan Yang; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 10857-10866

Abstract


Semi-supervised learning on class-imbalanced data, although a realistic problem, has been under studied. While existing semi-supervised learning (SSL) methods are known to perform poorly on minority classes, we find that they still generate high precision pseudo-labels on minority classes. By exploiting this property, in this work, we propose Class-Rebalancing Self-Training (CReST), a simple yet effective framework to improve existing SSL methods on class-imbalanced data. CReST iteratively retrains a baseline SSL model with a labeled set expanded by adding pseudo-labeled samples from an unlabeled set, where pseudo-labeled samples from minority classes are selected more frequently according to an estimated class distribution. We also propose a progressive distribution alignment to adaptively adjust the rebalancing strength dubbed CReST+. We show that CReST and CReST+ improve state-of-the-art SSL algorithms on various class-imbalanced datasets and consistently outperform other popular rebalancing methods. Code has been made available at https://github.com/google-research/crest.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Wei_2021_CVPR, author = {Wei, Chen and Sohn, Kihyuk and Mellina, Clayton and Yuille, Alan and Yang, Fan}, title = {CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learning}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2021}, pages = {10857-10866} }