A Unified Mixture-View Framework for Unsupervised Representation Learning
This work offers an orthogonal improvement for researchers and practitioners working on unsupervised contrastive representation learning, enhancing the performance of existing state-of-the-art methods.
This paper proposes a Beyond Single Instance Multi-view (BSIM) framework for unsupervised representation learning, which improves instance discrimination by measuring joint similarity between two sampled instances and their mixture. This approach leads to substantial performance gains across various downstream benchmarks like ImageNet-1k, PASCAL VOC 2007, and MS COCO 2017 when applied to existing methods like SimCLR, MoCo, and BYOL.
Recent unsupervised contrastive representation learning follows a Single Instance Multi-view (SIM) paradigm where positive pairs are usually constructed with intra-image data augmentation. In this paper, we propose an effective approach called Beyond Single Instance Multi-view (BSIM). Specifically, we impose more accurate instance discrimination capability by measuring the joint similarity between two randomly sampled instances and their mixture, namely spurious-positive pairs. We believe that learning joint similarity helps to improve the performance when encoded features are distributed more evenly in the latent space. We apply it as an orthogonal improvement for unsupervised contrastive representation learning, including current outstanding methods SimCLR, MoCo, and BYOL. We evaluate our learned representations on many downstream benchmarks like linear classification on ImageNet-1k and PASCAL VOC 2007, object detection on MS COCO 2017 and VOC, etc. We obtain substantial gains with a large margin almost on all these tasks compared with prior arts.