Novelty Detection Meets Collider Physics

TL;DR

This paper explores the application of autoencoder-based novelty detection to identify unknown physics signals in collider data, demonstrating high efficiency in recognizing new-physics events at LHC and future colliders.

Contribution

It introduces density-based novelty evaluators tailored for collider physics and addresses data fluctuation challenges, advancing model-independent detection methods.

Findings

High detection efficiency for new-physics signals at LHC.

Effective strategies to mitigate data fluctuation effects.

Potential for model-independent discovery of exotic particles.

Abstract

Novelty detection is the machine learning task to recognize data, which belong to an unknown pattern. Complementary to supervised learning, it allows to analyze data model-independently. We demonstrate the potential role of novelty detection in collider physics, using autoencoder-based deep neural network. Explicitly, we develop a set of density-based novelty evaluators, which are sensitive to the clustering of unknown-pattern testing data or new-physics signal events, for the design of detection algorithms. We also explore the influence of the known-pattern data fluctuations, arising from non-signal regions, on detection sensitivity. Strategies to address it are proposed. The algorithms are applied to detecting fermionic di-top partner and resonant di-top productions at LHC, and exotic Higgs decays of two specific modes at a $e^+e^-$ future collider. With parton-level analysis, we conclude that potentially the new-physics benchmarks can be recognized with high efficiency.