Graph theory inspired anomaly detection at the LHC

TL;DR

This paper introduces a graph autoencoder approach for anomaly detection in LHC data, leveraging graph theory to improve model interpretability and performance in high-dimensional collider event analysis.

Contribution

It develops a novel, model-agnostic graph autoencoder method using physically motivated sparse graph constructions for anomaly detection in collider data.

Findings

Best performance with intermediate subjet clustering and unique graph constructions

Graph connectivity influences jet classification accuracy

Graph-theoretic insights enhance interpretability of ML tools in collider physics

Abstract

Designing model-independent anomaly detection algorithms for analyzing LHC data remains a central challenge in the search for new physics, due to the high dimensionality of collider events. In this work, we develop a graph autoencoder as an unsupervised, model-agnostic tool for anomaly detection, using the LHC Olympics dataset as a benchmark. By representing jet constituents as a graph, we introduce a method to systematically control the information available to the model through sparse graph constructions that serve as physically motivated inductive biases. Specifically, (1) we construct graph autoencoders based on locally rigid Laman graphs and globally rigid unique graphs, and (2) we explore the clustering of jet constituents into subjets to interpolate between high- and low-level input representations. We obtain the best performance, measured in terms of the Significance Improvement Characteristic curve for an intermediate level of subjet clustering and certain sparse unique graph constructions. We further investigate the role of graph connectivity in jet classification tasks. Our results demonstrate the potential of leveraging graph-theoretic insights to refine and increase the interpretability of machine learning tools for collider experiments.