Subclass-balancing Contrastive Learning for Long-tailed Recognition

TL;DR

This paper introduces Subclass-balancing Contrastive Learning (SBCL), a novel method that improves long-tailed recognition by balancing subclasses within classes, capturing semantic substructures, and achieving state-of-the-art results.

Contribution

SBCL clusters head classes into subclasses to preserve semantic structures and balance representations, addressing limitations of existing long-tailed recognition methods.

Findings

SBCL achieves state-of-the-art performance on benchmark datasets.

It effectively balances instances and subclasses during training.

Extensive analyses confirm its advantages over existing methods.

Abstract

Long-tailed recognition with imbalanced class distribution naturally emerges in practical machine learning applications. Existing methods such as data reweighing, resampling, and supervised contrastive learning enforce the class balance with a price of introducing imbalance between instances of head class and tail class, which may ignore the underlying rich semantic substructures of the former and exaggerate the biases in the latter. We overcome these drawbacks by a novel ``subclass-balancing contrastive learning (SBCL)'' approach that clusters each head class into multiple subclasses of similar sizes as the tail classes and enforce representations to capture the two-layer class hierarchy between the original classes and their subclasses. Since the clustering is conducted in the representation space and updated during the course of training, the subclass labels preserve the semantic substructures of head classes. Meanwhile, it does not overemphasize tail class samples, so each individual instance contribute to the representation learning equally. Hence, our method achieves both the instance- and subclass-balance, while the original class labels are also learned through contrastive learning among subclasses from different classes. We evaluate SBCL over a list of long-tailed benchmark datasets and it achieves the state-of-the-art performance. In addition, we present extensive analyses and ablation studies of SBCL to verify its advantages.