Generalized Category Discovery via Reciprocal Learning and Class-Wise Distribution Regularization

TL;DR

This paper introduces RLCD, a novel framework for Generalized Category Discovery that combines reciprocal learning and class-wise distribution regularization to improve the identification of both base and novel classes efficiently.

Contribution

The paper proposes a reciprocal learning framework with class-wise distribution regularization, enhancing base discrimination and novel class recognition in GCD tasks.

Findings

RLCD outperforms existing methods on seven GCD datasets.

The auxiliary branch improves base class discrimination.

Class-wise distribution regularization boosts novel class performance.

Abstract

Generalized Category Discovery (GCD) aims to identify unlabeled samples by leveraging the base knowledge from labeled ones, where the unlabeled set consists of both base and novel classes. Since clustering methods are time-consuming at inference, parametric-based approaches have become more popular. However, recent parametric-based methods suffer from inferior base discrimination due to unreliable self-supervision. To address this issue, we propose a Reciprocal Learning Framework (RLF) that introduces an auxiliary branch devoted to base classification. During training, the main branch filters the pseudo-base samples to the auxiliary branch. In response, the auxiliary branch provides more reliable soft labels for the main branch, leading to a virtuous cycle. Furthermore, we introduce Class-wise Distribution Regularization (CDR) to mitigate the learning bias towards base classes. CDR essentially increases the prediction confidence of the unlabeled data and boosts the novel class performance. Combined with both components, our proposed method, RLCD, achieves superior performance in all classes with negligible extra computation. Comprehensive experiments across seven GCD datasets validate its superiority. Our codes are available at https://github.com/APORduo/RLCD.