What's the difference between OOD and Anomaly Detection?

[iCGY96]

What's the difference between OOD and Anomaly Detection?

From: #7

[iCGY96]

Out-of-distribution (OOD) detection is a classic and essential problem in machine learning, studied under different names, e.g., novelty or anomaly detection.

[1] V. Hodge and J. Austin. A survey of outlier detection methodologies. Artificial intelligence review, 2004.
[2] Jihoon Tack, Sangwoo Mo, Jongheon Jeong, Jinwoo Shin. CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances, NeurIPS 2020.