Semantic Positive Pairs for Enhancing Visual Representation Learning of Instance Discrimination methods

arXiv:2306.16122v2 Announce Type: replace-cross
Abstract: Self-supervised learning algorithms (SSL) based on instance discrimination have shown promising results, performing competitively or even outperforming supervised learning counterparts in some downstream tasks. Such approaches employ data augmentation to create two views of the same instance (i.e., positive pairs) and encourage the model to learn good representations by attracting these views closer in the embedding space without collapsing to the trivial solution. However, data augmentation is limited in representing positive pairs, and the repulsion …

abstract algorithms arxiv augmentation create cs.cv cs.lg data discrimination instance positive representation representation learning results self-supervised learning semantic ssl supervised learning tasks type visual