Robust Feature Learning Against Noisy Labels
This addresses the issue of noisy labels in supervised learning for deep neural networks, which is an incremental improvement in a domain-specific context.
The paper tackles the problem of noisy labels degrading deep neural network generalization by proposing an approach using unsupervised augmentation restoration, cluster regularization, and progressive self-bootstrapping to learn robust feature representations. The method enhances model robustness under severely noisy labels with minimal overhead.
Supervised learning of deep neural networks heavily relies on large-scale datasets annotated by high-quality labels. In contrast, mislabeled samples can significantly degrade the generalization of models and result in memorizing samples, further learning erroneous associations of data contents to incorrect annotations. To this end, this paper proposes an efficient approach to tackle noisy labels by learning robust feature representation based on unsupervised augmentation restoration and cluster regularization. In addition, progressive self-bootstrapping is introduced to minimize the negative impact of supervision from noisy labels. Our proposed design is generic and flexible in applying to existing classification architectures with minimal overheads. Experimental results show that our proposed method can efficiently and effectively enhance model robustness under severely noisy labels.