How does Disagreement Help Generalization against Label Corruption?
This addresses the issue of label corruption in weakly-supervised learning, offering a robust solution for training deep neural networks with noisy data, though it is incremental over prior methods like Co-teaching.
The paper tackles the problem of learning with noisy labels by proposing Co-teaching+, a method that prevents two networks from converging to consensus, leading to improved robustness in trained models, as demonstrated by superior empirical results on benchmark datasets.
Learning with noisy labels is one of the hottest problems in weakly-supervised learning. Based on memorization effects of deep neural networks, training on small-loss instances becomes very promising for handling noisy labels. This fosters the state-of-the-art approach "Co-teaching" that cross-trains two deep neural networks using the small-loss trick. However, with the increase of epochs, two networks converge to a consensus and Co-teaching reduces to the self-training MentorNet. To tackle this issue, we propose a robust learning paradigm called Co-teaching+, which bridges the "Update by Disagreement" strategy with the original Co-teaching. First, two networks feed forward and predict all data, but keep prediction disagreement data only. Then, among such disagreement data, each network selects its small-loss data, but back propagates the small-loss data from its peer network and updates its own parameters. Empirical results on benchmark datasets demonstrate that Co-teaching+ is much superior to many state-of-the-art methods in the robustness of trained models.