Null Hypothesis Test for Anomaly Detection

Abstract

We extend the use of Classification Without Labels for anomaly detection with a hypothesis test designed to exclude the background-only hypothesis. By testing for statistical independence of the two discriminating dataset regions, we are able to exclude the background-only hypothesis without relying on fixed anomaly score cuts or extrapolations of background estimates between regions. The method relies on the assumption of conditional independence of anomaly score features and dataset regions, which can be ensured using existing decorrelation techniques. As a benchmark example, we consider the LHC Olympics dataset where we show that mutual information represents a suitable test for statistical independence and our method exhibits excellent and robust performance at different signal fractions even in presence of realistic feature correlations.

10 pages, 3 figures, 1 Table. Matches published version at Physics Letters B. All code is available at https://github.com/ManuelSzewc/Null_Hypothesis_Test_for_Anomaly_Detection. Comments welcome!

References in corpus (2)

• Measuring and testing dependence by correlation of distances
• Learning new physics efficiently with nonparametric methods

Cited by in corpus (5)

• The Interplay of Machine Learning--based Resonant Anomaly Detection Methods
• Nanosecond anomaly detection with decision trees and real-time application to exotic Higgs decays
• Anomalies, Representations, and Self-Supervision
• Combining Resonant and Tail-based Anomaly Detection
• Non-resonant Anomaly Detection with Background Extrapolation