Enhancing Contrastive Learning with Efficient Combinatorial Positive Pairing
This work addresses the need for more efficient and effective unsupervised representation learning in computer vision, though it is incremental as it builds on existing multi-view methods.
The paper tackles the problem of improving contrastive learning by proposing a general multi-view strategy called ECPP, which enhances learning speed and performance, achieving state-of-the-art results on CIFAR-10 and ImageNet-100 benchmarks, with ECPP-boosted SimCLR outperforming supervised learning on ImageNet-100.
In the past few years, contrastive learning has played a central role for the success of visual unsupervised representation learning. Around the same time, high-performance non-contrastive learning methods have been developed as well. While most of the works utilize only two views, we carefully review the existing multi-view methods and propose a general multi-view strategy that can improve learning speed and performance of any contrastive or non-contrastive method. We first analyze CMC's full-graph paradigm and empirically show that the learning speed of $K$-views can be increased by $_{K}\mathrm{C}_{2}$ times for small learning rate and early training. Then, we upgrade CMC's full-graph by mixing views created by a crop-only augmentation, adopting small-size views as in SwAV multi-crop, and modifying the negative sampling. The resulting multi-view strategy is called ECPP (Efficient Combinatorial Positive Pairing). We investigate the effectiveness of ECPP by applying it to SimCLR and assessing the linear evaluation performance for CIFAR-10 and ImageNet-100. For each benchmark, we achieve a state-of-the-art performance. In case of ImageNet-100, ECPP boosted SimCLR outperforms supervised learning.