Vision Transformers and Graph Neural Networks for Charged Particle Tracking in the ATLAS Muon Spectrometer

TL;DR

This paper introduces machine learning approaches, including Graph Neural Networks and Vision Transformers, to improve charged particle tracking in the ATLAS Muon Spectrometer, addressing challenges posed by increased collision rates at the High-Luminosity LHC.

Contribution

It presents novel ML-based methods for background-hit rejection and end-to-end muon tracking, enhancing speed and efficiency in particle reconstruction.

Findings

Graph Neural Networks improve reconstruction speed by 15%.

Vision Transformers enable ultra-fast muon reconstruction in 2.3 ms.

Achieved 98% tracking efficiency with Transformer-based approach.

Abstract

The identification and reconstruction of charged particles, such as muons, is a main challenge for the physics program of the ATLAS experiment at the Large Hadron Collider. This task will become increasingly difficult with the start of the High-Luminosity LHC era after 2030, when the number of proton-proton collisions per bunch crossing will increase from 60 to up to 200. This elevated interaction density will also increase the occupancy within the ATLAS Muon Spectrometer, requiring more efficient and robust real-time data processing strategies within the experiment's trigger system, particularly the Event Filter. To address these algorithmic challenges, we present two machine-learning-based approaches. First, we target the problem of background-hit rejection in the Muon Spectrometer using Graph Neural Networks integrated into the non-ML baseline reconstruction chain, demonstrating a 15 % improvement in reconstruction speed (from 255 ms to 217 ms). Second, we present a proof-of-concept for end-to-end muon tracking using state-of-the-art Vision Transformer architectures, achieving ultra-fast approximate muon reconstruction in 2.3 ms on consumer-grade GPUs at 98 % tracking efficiency.