Learning from Noisy Labels with Contrastive Co-Transformer
This addresses the challenge of noisy labels in weakly supervised learning, which is a domain-specific problem for machine learning practitioners, and appears to be an incremental improvement based on the Co-Training framework.
The paper tackles the problem of deep learning with noisy labels by introducing a Contrastive Co-Transformer framework, which improves performance by a large margin over state-of-the-art methods on corrupted data from six benchmark datasets, including Clothing1M.
Deep learning with noisy labels is an interesting challenge in weakly supervised learning. Despite their significant learning capacity, CNNs have a tendency to overfit in the presence of samples with noisy labels. Alleviating this issue, the well known Co-Training framework is used as a fundamental basis for our work. In this paper, we introduce a Contrastive Co-Transformer framework, which is simple and fast, yet able to improve the performance by a large margin compared to the state-of-the-art approaches. We argue the robustness of transformers when dealing with label noise. Our Contrastive Co-Transformer approach is able to utilize all samples in the dataset, irrespective of whether they are clean or noisy. Transformers are trained by a combination of contrastive loss and classification loss. Extensive experimental results on corrupted data from six standard benchmark datasets including Clothing1M, demonstrate that our Contrastive Co-Transformer is superior to existing state-of-the-art methods.