Agreement or Disagreement in Noise-tolerant Mutual Learning?

TL;DR

This paper introduces MLC, a noise-tolerant deep learning framework that enhances robustness against noisy labels by using divergent regularization and agreement-based label correction, outperforming existing methods.

Contribution

The paper proposes a novel end-to-end noise-tolerant framework called MLC that employs divergent regularization and agreement-based label correction to improve learning with noisy labels.

Findings

MLC outperforms state-of-the-art methods on MNIST, CIFAR-10, and Clothing1M datasets.

MLC improves accuracy, generalization, and robustness in noisy label scenarios.

MLC is network-free and applicable to various tasks.

Abstract

Deep learning has made many remarkable achievements in many fields but suffers from noisy labels in datasets. The state-of-the-art learning with noisy label method Co-teaching and Co-teaching+ confronts the noisy label by mutual-information between dual-network. However, the dual network always tends to convergent which would weaken the dual-network mechanism to resist the noisy labels. In this paper, we proposed a noise-tolerant framework named MLC in an end-to-end manner. It adjusts the dual-network with divergent regularization to ensure the effectiveness of the mechanism. In addition, we correct the label distribution according to the agreement between dual-networks. The proposed method can utilize the noisy data to improve the accuracy, generalization, and robustness of the network. We test the proposed method on the simulate noisy dataset MNIST, CIFAR-10, and the real-world noisy dataset Clothing1M. The experimental result shows that our method outperforms the previous state-of-the-art method. Besides, our method is network-free thus it is applicable to many tasks. Our code can be found at https://github.com/JiarunLiu/MLC.