Learning representations of irregular particle-detector geometry with distance-weighted graph networks

Abstract

We explore the use of graph networks to deal with irregular-geometry detectors in the context of particle reconstruction. Thanks to their representation-learning capabilities, graph networks can exploit the full detector granularity, while natively managing the event sparsity and arbitrarily complex detector geometries. We introduce two distance-weighted graph network architectures, dubbed GarNet and GravNet layers, and apply them to a typical particle reconstruction task. The performance of the new architectures is evaluated on a data set of simulated particle interactions on a toy model of a highly granular calorimeter, loosely inspired by the endcap calorimeter to be installed in the CMS detector for the High-Luminosity LHC phase. We study the clustering of energy depositions, which is the basis for calorimetric particle reconstruction, and provide a quantitative comparison to alternative approaches. The proposed algorithms provide an interesting alternative to existing methods, offering equally performing or less resource-demanding solutions with less underlying assumptions on the detector geometry and, consequently, the possibility to generalize to other detectors.

10p, v2: published version

References in corpus (6)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Particle-flow reconstruction and global event description with the CMS detector
• ParticleNet: Jet Tagging via Particle Clouds
• Energy Flow Networks: Deep Sets for Particle Jets
• Pileup Mitigation with Machine Learning (PUMML)
• Studies of Boosted Decision Trees for MiniBooNE Particle Identification

Cited by in corpus (65)

• Graph Neural Networks in Particle Physics
• Getting High: High Fidelity Simulation of High Granularity Calorimeters with High Speed
• JEDI-net: a jet identification algorithm based on interaction networks
• ABCNet: An attention-based method for particle tagging
• Graph Neural Networks for Particle Reconstruction in High Energy Physics detectors
• Applications and Techniques for Fast Machine Learning in Science
• MLPF: Efficient machine-learned particle-flow reconstruction using graph neural networks
• Interaction networks for the identification of boosted $H\to b\overline{b}$ decays
• ABCDisCo: Automating the ABCD Method with Machine Learning
• CaloClouds: Fast Geometry-Independent Highly-Granular Calorimeter Simulation
• A guide for deploying Deep Learning in LHC searches: How to achieve optimality and account for uncertainty
• Casting a graph net to catch dark showers
• Point Cloud Transformers applied to Collider Physics
• Charged particle tracking via edge-classifying interaction networks
• Supervised Jet Clustering with Graph Neural Networks for Lorentz Boosted Bosons
• EPiC-GAN: Equivariant Point Cloud Generation for Particle Jets
• Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics
• Symmetries, Safety, and Self-Supervision
• Object condensation: one-stage grid-free multi-object reconstruction in physics detectors, graph and image data
• Graph Neural Networks for Charged Particle Tracking on FPGAs
• Towards a Computer Vision Particle Flow
• Neural Network-based Top Tagger with Two-Point Energy Correlations and Geometry of Soft Emissions
• On the Use of Neural Networks for Energy Reconstruction in High-granularity Calorimeters
• Reconstructing particles in jets using set transformer and hypergraph prediction networks
• Shared Data and Algorithms for Deep Learning in Fundamental Physics
• Probing triple Higgs coupling with machine learning at the LHC
• SLAPS: Self-Supervision Improves Structure Learning for Graph Neural Networks
• LL-GNN: Low Latency Graph Neural Networks on FPGAs for High Energy Physics
• Machine Learning for Particle Flow Reconstruction at CMS
• Jet tagging algorithm of graph network with HaarPooling message passing
• GPU coprocessors as a service for deep learning inference in high energy physics
• A high-granularity calorimeter insert based on SiPM-on-tile technology at the future Electron-Ion Collider
• TomOpt: Differential optimisation for task- and constraint-aware design of particle detectors in the context of muon tomography
• Photon Reconstruction in the Belle II Calorimeter Using Graph Neural Networks
• Spatio-Temporal Anomaly Detection with Graph Networks for Data Quality Monitoring of the Hadron Calorimeter
• Graph neural network for 3D classification of ambiguities and optical crosstalk in scintillator-based neutrino detectors
• Graph Generative Models for Fast Detector Simulations in High Energy Physics
• Fast convolutional neural networks for identifying long-lived particles in a high-granularity calorimeter
• A Tailored Convolutional Neural Network for Nonlinear Manifold Learning of Computational Physics Data using Unstructured Spatial Discretizations
• Real-time Graph Building on FPGAs for Machine Learning Trigger Applications in Particle Physics
• Semi-supervised Graph Neural Networks for Pileup Noise Removal
• Calorimetric Measurement of Multi-TeV Muons via Deep Regression
• Optimising longitudinal and lateral calorimeter granularity for software compensation in hadronic showers using deep neural networks
• Table Structure Recognition using Top-Down and Bottom-Up Cues
• Reconstruction of decays to merged photons using end-to-end deep learning with domain continuation in the CMS detector
• Portable acceleration of CMS computing workflows with coprocessors as a service
• Role of polarizations and spin-spin correlations of $W^{\prime }s$ in $e^{-}{e}^{+}\to W^{-}{W}^{+}$ at $\sqrt{s}=250$ GeV to probe anomalous $W^{-}{W}^{+}Z/\gamma$ couplings
• End-to-End Multi-Track Reconstruction using Graph Neural Networks at Belle II
• Partition Pooling for Convolutional Graph Network Applications in Particle Physics
• Leveraging Staggered Tessellation for Enhanced Spatial Resolution in High-Granularity Calorimeters
• Progress in End-to-End Optimization of Detectors for Fundamental Physics with Differentiable Programming
• A Dynamic Reduction Network for Point Clouds
• Foundations of automatic feature extraction at LHC--point clouds and graphs
• Performance of a Geometric Deep Learning Pipeline for HL-LHC Particle Tracking
• Particle Graph Autoencoders and Differentiable, Learned Energy Mover's Distance
• Scalable Multi-Task Learning for Particle Collision Event Reconstruction with Heterogeneous Graph Neural Networks
• Particle Track Reconstruction using Geometric Deep Learning
• Explaining machine-learned particle-flow reconstruction
• Reconstruction of electromagnetic showers in calorimeters using Deep Learning
• How to use Machine Learning to improve the discrimination between signal and background at particle colliders
• Optimised Graph Convolution for Calorimetry Event Classification
• Neutral pion reconstruction using machine learning in the MINERvA experiment at $⟨E_{\nu }⟩\sim 6$ GeV
• Real-time graph neural networks on FPGAs for the Belle II electromagnetic calorimeter
• End-to-end multi-particle reconstruction in high occupancy imaging calorimeters with graph neural networks
• Convolutions for Spatial Interaction Modeling