Robust Collaborative Learning with Noisy Labels
This work is significant for researchers and practitioners working with noisy labeled datasets, as it offers an improved method for robust learning by mitigating sample selection bias in curriculum learning.
This paper addresses the problem of learning with noisy labels, where existing curriculum learning methods suffer from sample selection bias. The authors propose Robust Collaborative Learning (RCL), a novel framework that leverages both disagreement and agreement between multiple networks to reduce noise in gradients, demonstrating effectiveness on synthetic image data and real-world bioinformatics data.
Learning with curriculum has shown great effectiveness in tasks where the data contains noisy (corrupted) labels, since the curriculum can be used to re-weight or filter out noisy samples via proper design. However, obtaining curriculum from a learner itself without additional supervision or feedback deteriorates the effectiveness due to sample selection bias. Therefore, methods that involve two or more networks have been recently proposed to mitigate such bias. Nevertheless, these studies utilize the collaboration between networks in a way that either emphasizes the disagreement or focuses on the agreement while ignores the other. In this paper, we study the underlying mechanism of how disagreement and agreement between networks can help reduce the noise in gradients and develop a novel framework called Robust Collaborative Learning (RCL) that leverages both disagreement and agreement among networks. We demonstrate the effectiveness of RCL on both synthetic benchmark image data and real-world large-scale bioinformatics data.