Reviewed Version (pdf): https://openreview.net/references/pdf?id=Gv5zNv5j8
Keywords: anomaly detection, out-of-distribution detection, OOD detection, outlier detection, density estimation
Abstract: Thanks to the tractability of their likelihood, some deep generative models show promise for seemingly straightforward but important applications like anomaly detection, uncertainty estimation, and active learning. However, the likelihood values empirically attributed to anomalies conflict with the expectations these proposed applications suggest. In this paper, we take a closer look at the behavior of distribution densities and show that these quantities carry less meaningful information than previously thought, beyond estimation issues or the curse of dimensionality. We conclude that the use of these likelihoods for out-of-distribution detection relies on strong and implicit hypotheses and highlight the necessity of explicitly formulating these assumptions for reliable anomaly detection.
One-sentence Summary: Explaining issues of density models for anomaly detection.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics