Cooperative Learning for Noisy Supervision

TL;DR

This paper introduces the Cooperative Learning (CooL) framework that uses dual networks to improve learning with noisy labels, providing theoretical insights and demonstrating superior performance on various benchmarks.

Contribution

It offers the first theoretical explanation for the benefits of dual networks in noisy supervision and proposes an effective combination method within CooL.

Findings

CooL outperforms state-of-the-art methods on multiple benchmarks.

The framework provides a more reliable risk minimization for clean data.

Experimental results validate the effectiveness of the theoretical analysis.

Abstract

Learning with noisy labels has gained the enormous interest in the robust deep learning area. Recent studies have empirically disclosed that utilizing dual networks can enhance the performance of single network but without theoretic proof. In this paper, we propose Cooperative Learning (CooL) framework for noisy supervision that analytically explains the effects of leveraging dual or multiple networks. Specifically, the simple but efficient combination in CooL yields a more reliable risk minimization for unseen clean data. A range of experiments have been conducted on several benchmarks with both synthetic and real-world settings. Extensive results indicate that CooL outperforms several state-of-the-art methods.