Performance of a Geometric Deep Learning Pipeline for HL-LHC Particle Tracking

Abstract

The Exa.TrkX project has applied geometric learning concepts such as metric learning and graph neural networks to HEP particle tracking. Exa.TrkX's tracking pipeline groups detector measurements to form track candidates and filters them. The pipeline, originally developed using the TrackML dataset (a simulation of an LHC-inspired tracking detector), has been demonstrated on other detectors, including DUNE Liquid Argon TPC and CMS High-Granularity Calorimeter. This paper documents new developments needed to study the physics and computing performance of the Exa.TrkX pipeline on the full TrackML dataset, a first step towards validating the pipeline using ATLAS and CMS data. The pipeline achieves tracking efficiency and purity similar to production tracking algorithms. Crucially for future HEP applications, the pipeline benefits significantly from GPU acceleration, and its computational requirements scale close to linearly with the number of particles in the event.

References in corpus (21)

• Geometric deep learning: going beyond Euclidean data
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• The Effectiveness of Data Augmentation in Image Classification using Deep Learning
• Fast Graph Representation Learning with PyTorch Geometric
• Description and performance of track and primary-vertex reconstruction with the CMS tracker
• Measuring the leading hadronic contribution to the muon g-2 via $\mu e$ scattering
• Performance of the ATLAS Track Reconstruction Algorithms in Dense Environments in LHC Run 2
• Learning representations of irregular particle-detector geometry with distance-weighted graph networks
• Accelerating 3D Deep Learning with PyTorch3D
• Graph Neural Networks for Particle Reconstruction in High Energy Physics detectors
• Dissecting the Graphcore IPU Architecture via Microbenchmarking
• The Tracking Machine Learning challenge : Accuracy phase
• Novel deep learning methods for track reconstruction
• Accelerated Charged Particle Tracking with Graph Neural Networks on FPGAs
• TFLMS: Large Model Support in TensorFlow by Graph Rewriting
• Distributed Training of Graph Convolutional Networks
• Overhaul and Installation of the ICARUS-T600 Liquid Argon TPC Electronics for the FNAL Short Baseline Neutrino Program
• GPU coprocessors as a service for deep learning inference in high energy physics
• Graph Neural Network for Object Reconstruction in Liquid Argon Time Projection Chambers
• GPU-based Online Track Reconstruction for the ALICE TPC in Run 3 with Continuous Read-Out
• Comparison of two hardware-based hit filtering methods for trackers in high-pileup environments