Does Lorentz-symmetric design boost network performance in jet physics?

Abstract

In the deep learning era, improving the neural network performance in jet physics is a rewarding task as it directly contributes to more accurate physics measurements at the LHC. Recent research has proposed various network designs in consideration of the full Lorentz symmetry, but its benefit is still not systematically asserted, given that there remain many successful networks without taking it into account. We conduct a detailed study on the Lorentz-symmetric design. We propose two generalized approaches for modifying a network - these methods are experimented on Particle Flow Network, ParticleNet, and LorentzNet, and exhibit a general performance gain. We also reveal that the notable improvement attributed to the "pairwise mass" feature in the network is due to its introduction of a structure that fully complies with Lorentz symmetry. We confirm that Lorentz-symmetry preservation serves as a strong inductive bias of jet physics, hence calling for attention to such general recipes in future network designs.

arXiv comment16 pages, 7 figures

References in corpus (31)

• The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
• The anti-k_t jet clustering algorithm
• An Introduction to PYTHIA 8.2
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• ParticleNet: Jet Tagging via Particle Clouds
• ATLAS $b$-jet identification performance and efficiency measurement with $t\bar{t}$ events in $pp$ collisions at $\sqrt{s}=13$ TeV
• Jet-Images -- Deep Learning Edition
• Energy Flow Networks: Deep Sets for Particle Jets
• Deep-learned Top Tagging with a Lorentz Layer
• Jet-Images: Computer Vision Inspired Techniques for Jet Tagging
• Graph Neural Networks in Particle Physics
• Identification of heavy, energetic, hadronically decaying particles using machine-learning techniques
• JEDI-net: a jet identification algorithm based on interaction networks
• An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging
• ABCNet: An attention-based method for particle tagging
• Anomaly detection with Convolutional Graph Neural Networks
• MLPF: Efficient machine-learned particle-flow reconstruction using graph neural networks
• 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
• Lorentz Boost Networks: Autonomous Physics-Inspired Feature Engineering
• Deep Set Auto Encoders for Anomaly Detection in Particle Physics
• Permutationless Many-Jet Event Reconstruction with Symmetry Preserving Attention Networks
• The Boosted Higgs Jet Reconstruction via Graph Neural Network
• A deep neural network for simultaneous estimation of b jet energy and resolution
• Energy-weighted Message Passing: an infra-red and collinear safe graph neural network algorithm
• Search for Higgs boson decay to a charm quark-antiquark pair in proton-proton collisions at $\sqrt{s}$ = 13 TeV
• Lorentz group equivariant autoencoders
• Jet tagging algorithm of graph network with HaarPooling message passing
• A Holistic Approach to Predicting Top Quark Kinematic Properties with the Covariant Particle Transformer
• BIP: Boost Invariant Polynomials for Efficient Jet Tagging

Cited by in corpus (6)

• Equivariant, Safe and Sensitive -- Graph Networks for New Physics
• Accuracy versus precision in boosted top tagging with the ATLAS detector
• Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments
• Foundations of automatic feature extraction at LHC--point clouds and graphs
• Learning Broken Symmetries with Approximate Invariance
• jBOT: Semantic Jet Representation Clustering Emerges from Self-Distillation