Understanding Negative Samples in Instance Discriminative Self-supervised Representation Learning. (arXiv:2102.06866v5 [cs.LG] UPDATED)

Instance discriminative self-supervised representation learning has been
attracted attention thanks to its unsupervised nature and informative feature
representation for downstream tasks. In practice, it commonly uses a larger
number of negative samples than the number of supervised classes. However,
there is an inconsistency in the existing analysis; theoretically, a large
number of negative samples degrade classification performance on a downstream
supervised task, while empirically, they improve the performance. We provide a
novel framework to analyze this empirical result regarding negative …

arxiv learning negative