Perfect Density Models Cannot Guarantee Anomaly Detection

Citations Over TimeTop 10% of 2021 papers

Abstract

Thanks to the tractability of their likelihood, several 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 through the lens of reparametrization 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 anomaly detection relies on strong and implicit hypotheses, and highlight the necessity of explicitly formulating these assumptions for reliable anomaly detection.

Related Papers

• → Explainable Anomaly Detection Framework for Maritime Main Engine Sensor Data(2021)59 cited
• → Towards Experienced Anomaly Detector Through Reinforcement Learning(2018)56 cited
• → Anomaly Detection with Partially Observed Anomaly Types(2021)4 cited
• → Human-machine interactive streaming anomaly detection by online self-adaptive forest(2022)8 cited
• → Tree-based Self-adaptive Anomaly Detection by Human-Machine Interaction(2021)1 cited