Self-supervised Anomaly Detection for New Physics

TL;DR

This paper introduces a self-supervised, model-agnostic method using transformer neural networks to encode particle collision data into a low-dimensional space for efficient anomaly detection of new physics signals.

Contribution

It demonstrates the effectiveness of using a contrastively trained transformer to create latent space representations for anomaly detection in high-energy physics.

Findings

Latent space classifier performance is comparable to event space classifier.

Weakly supervised bump hunt in latent space matches physical space detection.

Latent representations enable computationally efficient anomaly searches.

Abstract

We investigate a method of model-agnostic anomaly detection through studying jets, collimated sprays of particles produced in high-energy collisions. We train a transformer neural network to encode simulated QCD "event space" dijets into a low-dimensional "latent space" representation. We optimize the network using the self-supervised contrastive loss, which encourages the preservation of known physical symmetries of the dijets. We then train a binary classifier to discriminate a BSM resonant dijet signal from a QCD dijet background both in the event space and the latent space representations. We find the classifier performances on the event and latent spaces to be comparable. We finally perform an anomaly detection search using a weakly supervised bump hunt on the latent space dijets, finding again a comparable performance to a search run on the physical space dijets. This opens the door to using low-dimensional latent representations as a computationally efficient space for resonant anomaly detection in generic particle collision events.