Understanding normalization in contrastive representation learning and out-of-distribution detection

arXiv:2312.15288v2 Announce Type: replace
Abstract: Contrastive representation learning has emerged as an outstanding approach for anomaly detection. In this work, we explore the $\ell_2$-norm of contrastive features and its applications in out-of-distribution detection. We propose a simple method based on contrastive learning, which incorporates out-of-distribution data by discriminating against normal samples in the contrastive layer space. Our approach can be applied flexibly as an outlier exposure (OE) approach, where the out-of-distribution data is a huge collective of random images, or …

abstract anomaly anomaly detection applications arxiv cs.cv data detection distribution explore features norm normal normalization representation representation learning samples simple stat.ml type understanding work