A Simple yet Effective Baseline for Robust Deep Learning with Noisy Labels

Abstract

Recently deep neural networks have shown their capacity to memorize training data, even with noisy labels, which hurts generalization performance. To mitigate this issue, we propose a simple but effective method that is robust to noisy labels, even with severe noise. Our objective involves a variance regularization term that implicitly penalizes the Jacobian norm of the neural network on the whole training set (including the noisy-labeled data), which encourages generalization and prevents overfitting to the corrupted labels. Experiments on noisy benchmarks demonstrate that our approach achieves state-of-the-art performance with a high tolerance to severe noise.

Tasks

Reproductions