Unsupervised Hadronic SUEP at the LHC

TL;DR

This paper explores methods to detect Soft Unclustered Energy Patterns (SUEP) from confining dark sectors at the LHC, focusing on prompt dark hadron decays in exotic Higgs processes, using novel observables and machine learning techniques.

Contribution

It introduces new observables and an unsupervised autoencoder approach for SUEP detection, enhancing sensitivity without relying on detailed signal modeling.

Findings

HL-LHC can probe exotic Higgs decays to dark sectors at percent-level branching ratios.

Proposed observables effectively distinguish SUEP events from background.

Unsupervised methods are robust and broadly applicable to various SUEP searches.

Abstract

Confining dark sectors with pseudo-conformal dynamics produce SUEP, or Soft Unclustered Energy Patterns, at colliders: isotropic dark hadrons with soft and democratic energies. We target the experimental nightmare scenario, SUEPs in exotic Higgs decays, where all dark hadrons decay promptly to SM hadrons. First, we identify three promising observables, the charged particle multiplicity, the event ring isotropy, and the matrix of geometric distances between charged tracks. Their patterns can be exploited through a cut-and-count search, supervised machine learning, or an unsupervised autoencoder. We find that the HL-LHC will probe exotic Higgs branching ratios at the per-cent level, even without a detailed knowledge of the signal features. Our techniques can be applied to other SUEP searches, especially the unsupervised strategy, which is independent of overly specific model assumptions and the corresponding precision simulations.