Robust Anomaly Detection for Particle Physics Using Multi-Background Representation Learning

TL;DR

This paper introduces a robust anomaly detection method for particle physics that leverages multi-background representation learning and decorrelation to improve detection of new particles in collider data.

Contribution

It proposes a novel multi-background representation learning approach and generalized decorrelation for enhanced anomaly detection in high-energy physics.

Findings

Improved detection accuracy using multi-background representation.

Enhanced robustness through generalized decorrelation.

Validated on Large Hadron Collider data.

Abstract

Anomaly, or out-of-distribution, detection is a promising tool for aiding discoveries of new particles or processes in particle physics. In this work, we identify and address two overlooked opportunities to improve anomaly detection for high-energy physics. First, rather than train a generative model on the single most dominant background process, we build detection algorithms using representation learning from multiple background types, thus taking advantage of more information to improve estimation of what is relevant for detection. Second, we generalize decorrelation to the multi-background setting, thus directly enforcing a more complete definition of robustness for anomaly detection. We demonstrate the benefit of the proposed robust multi-background anomaly detection algorithms on a high-dimensional dataset of particle decays at the Large Hadron Collider.