Deep sets and event-level maximum-likelihood estimation for fast pile-up jet rejection in ATLAS

TL;DR

This paper introduces a novel deep learning approach using Deep Sets and maximum likelihood estimation to efficiently reject pile-up jets in real-time at the ATLAS trigger system, improving multi-jet event selection at high luminosities.

Contribution

The paper presents DIPz, a new uncertainty-aware jet regression model based on Deep Sets, and an event-level discriminant MLPL for pile-up rejection in high-luminosity LHC conditions.

Findings

DIPz effectively regresses jet origin positions using charged particle tracks.

MLPL improves event selection for multi-jet signatures.

The combined method is computationally efficient for real-time trigger applications.

Abstract

Multiple proton-proton collisions (pile-up) occur at every bunch crossing at the LHC, with the mean number of interactions expected to reach 80 during Run 3 and up to 200 at the High-Luminosity LHC. As a direct consequence, events with multijet signatures will occur at increasingly high rates. To cope with the increased luminosity, being able to efficiently group jets according to their origin along the beamline is crucial, particularly at the trigger level. In this work, a novel uncertainty-aware jet regression model based on a Deep Sets architecture is introduced, DIPz, to regress on a jet origin position along the beamline. The inputs to the DIPz algorithm are the charged particle tracks associated to each jet. An event-level discriminant, the Maximum Log Product of Likelihoods (MLPL), is constructed by combining the DIPz per-jet predictions. MLPL is cut-optimized to select events compatible with targeted multi-jet signature selection. This combined approach provides a robust and computationally efficient method for pile-up rejection in multi-jet final states, applicable to real-time event selections at the ATLAS High Level Trigger.