Hypergraphs in LHC Phenomenology -- The Next Frontier of IRC-Safe Feature Extraction

Abstract

In this study, we critically evaluate the approximation capabilities of existing infra-red and collinear (IRC) safe feature extraction algorithms, namely Energy Flow Networks (EFNs) and Energy-weighted Message Passing Networks (EMPNs). Our analysis reveals that these algorithms fall short in extracting features from any $N$-point correlation that isn't a power of two, based on the complete basis of IRC safe observables, specifically C-correlators. To address this limitation, we introduce the Hypergraph Energy-weighted Message Passing Networks (H-EMPNs), designed to capture any $N$-point correlation among particles efficiently. Using the case study of top vs. QCD jets, which holds significant information in its 3-point correlations, we demonstrate that H-EMPNs targeting up to N=3 correlations exhibit superior performance compared to EMPNs focusing on up to N=4 correlations within jet constituents.

arXiv commentMinor modifications in text and figure. Matches published version

References in corpus (34)

• The anti-k_t jet clustering algorithm
• An Introduction to PYTHIA 8.2
• FastJet user manual
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• ParticleNet: Jet Tagging via Particle Clouds
• Energy Flow Networks: Deep Sets for Particle Jets
• OmniFold: A Method to Simultaneously Unfold All Observables
• Energy flow polynomials: A complete linear basis for jet substructure
• JEDI-net: a jet identification algorithm based on interaction networks
• The Metric Space of Collider Events
• An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• ABCNet: An attention-based method for particle tagging
• Anomaly detection with Convolutional Graph Neural Networks
• Understanding Event-Generation Networks via Uncertainties
• Measuring Multijet Structure of Hadronic Energy Flow Or What IS A Jet?
• Generative Networks for Precision Enthusiasts
• A factorisation-aware Matrix element emulator
• Deep-Learning Jets with Uncertainties and More
• Casting a graph net to catch dark showers
• Uncertainty Aware Learning for High Energy Physics
• Mapping Machine-Learned Physics into a Human-Readable Space
• Using neural networks for efficient evaluation of high multiplicity scattering amplitudes
• The top quark legacy of the LHC Run II for PDF and SMEFT analyses
• Measuring QCD Splittings with Invertible Networks
• Topological Obstructions to Autoencoding
• Energy-weighted Message Passing: an infra-red and collinear safe graph neural network algorithm
• Infrared Safety of a Neural-Net Top Tagging Algorithm
• Towards Machine Learning Analytics for Jet Substructure
• Lorentz group equivariant autoencoders
• Comparing Machine Learning and Interpolation Methods for Loop-Level Calculations
• Improved Constraints on Effective Top Quark Interactions using Edge Convolution Networks
• Neural Embedding: Learning the Embedding of the Manifold of Physics Data
• Comparing Point Cloud Strategies for Collider Event Classification

Cited by in corpus (6)

• Equivariant, Safe and Sensitive -- Graph Networks for New Physics
• Safe but Incalculable: Energy-weighting is not all you need
• Interplay of Traditional Methods and Machine Learning Algorithms for Tagging Boosted Objects
• Foundations of automatic feature extraction at LHC--point clouds and graphs
• Reconstructing short-lived particles using hypergraph representation learning
• Stable and Interpretable Jet Physics with IRC-Safe Equivariant Feature Extraction