Unsupervised in-distribution anomaly detection of new physics through conditional density estimation

TL;DR

This paper introduces an unsupervised in-distribution anomaly detection method using conditional density estimation, successfully identifying a new particle in LHC collision data with state-of-the-art performance.

Contribution

The paper presents a novel unsupervised in-distribution anomaly detection approach based on conditional density estimation, applied to high-energy physics data.

Findings

Detected a new particle in 0.08% of collision events

Achieved state-of-the-art performance in the 2020 LHC Olympics

Demonstrated effectiveness of in-distribution anomaly detection

Abstract

Anomaly detection is a key application of machine learning, but is generally focused on the detection of outlying samples in the low probability density regions of data. Here we instead present and motivate a method for unsupervised in-distribution anomaly detection using a conditional density estimator, designed to find unique, yet completely unknown, sets of samples residing in high probability density regions. We apply this method towards the detection of new physics in simulated Large Hadron Collider (LHC) particle collisions as part of the 2020 LHC Olympics blind challenge, and show how we detected a new particle appearing in only 0.08% of 1 million collision events. The results we present are our original blind submission to the 2020 LHC Olympics, where it achieved the state-of-the-art performance.