Contrastive Learning with Stronger Augmentations
This work addresses a bottleneck in representation learning for computer vision by enabling the use of stronger augmentations, though it is incremental as it builds upon existing contrastive learning methods.
The paper tackles the challenge of using strong data augmentations in contrastive learning, which often distort image structures and hinder retrieval, by proposing CLSA, a framework that leverages distribution divergence to improve performance. It achieves a top-1 accuracy of 76.2% on ImageNet with a ResNet-50, nearly matching supervised results of 76.5%.
Representation learning has significantly been developed with the advance of contrastive learning methods. Most of those methods have benefited from various data augmentations that are carefully designated to maintain their identities so that the images transformed from the same instance can still be retrieved. However, those carefully designed transformations limited us to further explore the novel patterns exposed by other transformations. Meanwhile, as found in our experiments, the strong augmentations distorted the images' structures, resulting in difficult retrieval. Thus, we propose a general framework called Contrastive Learning with Stronger Augmentations~(CLSA) to complement current contrastive learning approaches. Here, the distribution divergence between the weakly and strongly augmented images over the representation bank is adopted to supervise the retrieval of strongly augmented queries from a pool of instances. Experiments on the ImageNet dataset and downstream datasets showed the information from the strongly augmented images can significantly boost the performance. For example, CLSA achieves top-1 accuracy of 76.2% on ImageNet with a standard ResNet-50 architecture with a single-layer classifier fine-tuned, which is almost the same level as 76.5% of supervised results. The code and pre-trained models are available in https://github.com/maple-research-lab/CLSA.