RenyiCL: Contrastive Representation Learning with Skew Renyi Divergence

Contrastive representation learning seeks to acquire useful representations by estimating the shared information between multiple views of data. Here, the choice of data augmentation is sensitive to the quality of learned representations: as harder the data augmentations are applied, the views share more task-relevant information, but also task-irrelevant one that can hinder the generalization capability of representation. Motivated by this, we present a new robust contrastive learning scheme, coined RenyiCL, which can effectively manage harder augmentations by utilizing Renyi divergence. Our method is built upon the variational lower bound of Renyi divergence, but a naive usage of a variational method is impractical due to the large variance. To tackle this challenge, we propose a novel contrastive objective that conducts variational estimation of a skew Renyi divergence and provide a theoretical guarantee on how variational estimation of skew divergence leads to stable training. We show that Renyi contrastive learning objectives perform innate hard negative sampling and easy positive sampling simultaneously so that it can selectively learn useful features and ignore nuisance features. Through experiments on ImageNet, we show that Renyi contrastive learning with stronger augmentations outperforms other self-supervised methods without extra regularization or computational overhead. Moreover, we also validate our method on other domains such as graph and tabular, showing empirical gain over other contrastive methods. The implementation and pre-trained models are available at (1).