The Mass-ive Issue: Anomaly Detection in Jet Physics

TL;DR

This paper explores the challenges of applying anomaly detection techniques, specifically Variational Autoencoders, to jet physics in particle experiments, emphasizing the importance of unbiased jet mass estimation.

Contribution

It identifies the critical challenge of maintaining unbiased jet mass estimates when using machine learning-based anomaly detection in particle physics.

Findings

Unavoidable bias in jet mass estimation with current anomaly detection methods.

Variational Autoencoders demonstrate limitations in unbiased anomaly detection.

Highlights the need for new approaches to preserve jet mass integrity.

Abstract

In the hunt for new and unobserved phenomena in particle physics, attention has turned in recent years to using advanced machine learning techniques for model independent searches. In this paper we highlight the main challenge of applying anomaly detection to jet physics, where preserving an unbiased estimator of the jet mass remains a critical piece of any model independent search. Using Variational Autoencoders and multiple industry-standard anomaly detection metrics, we demonstrate the unavoidable nature of this problem.