Dynamics-Aware Loss for Learning with Label Noise

Label noise poses a serious threat to deep neural networks (DNNs). Employing robust loss function which reconciles fitting ability with robustness is a simple but effective strategy to handle this problem. However, the widely-used static trade-off between these two factors contradicts the dynamic nature of DNNs learning with label noise, leading to inferior performance. Therefore, we propose a dynamics-aware loss (DAL) to solve this problem. Considering that DNNs tend to first learn generalized patterns, then gradually overfit label noise, DAL strengthens the fitting ability initially, then gradually increases the weight of robustness. Moreover, at the later stage, we let DNNs put more emphasis on easy examples which are more likely to be correctly labeled than hard ones and introduce a bootstrapping term to further reduce the negative impact of label noise. Both the detailed theoretical analyses and extensive experimental results demonstrate the superiority of our method.