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

By Chen Wei, Kihyuk Sohn, Clayton Mellina, Alan L. Yuille, Fan Yang. Computer Vision and Pattern Recognition, 2021

337 citations recorded by the source corpus.

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 pseudolabeled samples from an unlabeled set, where pseudolabeled 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.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering, Media Technology

View Paper: https://www.semanticscholar.org/paper/09acced5fcb49322f5a26ac7a4cbe9f1308657c4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Electrical and Electronic Engineering

Luc Van Gool

In Media Technology

Citations in the corpus

DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation (2021, background): “sampling [23,25,81], loss re-weighting [42,59], and transfer learning [36, 43].”