Lorentz group equivariant autoencoders

Abstract

There has been significant work recently in developing machine learning (ML) models in high energy physics (HEP) for tasks such as classification, simulation, and anomaly detection. Often these models are adapted from those designed for datasets in computer vision or natural language processing, which lack inductive biases suited to HEP data, such as equivariance to its inherent symmetries. Such biases have been shown to make models more performant and interpretable, and reduce the amount of training data needed. To that end, we develop the Lorentz group autoencoder (LGAE), an autoencoder model equivariant with respect to the proper, orthochronous Lorentz group ${\mathrm{SO}}^{+}(3,1)$, with a latent space living in the representations of the group. We present our architecture and several experimental results on jets at the LHC and find it outperforms graph and convolutional neural network baseline models on several compression, reconstruction, and anomaly detection metrics. We also demonstrate the advantage of such an equivariant model in analyzing the latent space of the autoencoder, which can improve the explainability of potential anomalies discovered by such ML models.

arXiv comment11 pages, 7 figures, 4 tables, and a 3 page appendix

References in corpus (40)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• 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
• Machine learning and the physical sciences
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Searching for Exotic Particles in High-Energy Physics with Deep Learning
• ParticleNet: Jet Tagging via Particle Clouds
• Deep Learning and its Application to LHC Physics
• CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks
• Jet Substructure at the Large Hadron Collider: A Review of Recent Advances in Theory and Machine Learning
• Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis
• Accelerating Science with Generative Adversarial Networks: An Application to 3D Particle Showers in Multi-Layer Calorimeters
• Searching for New Physics with Deep Autoencoders
• Energy Flow Networks: Deep Sets for Particle Jets
• Deep-learned Top Tagging with a Lorentz Layer
• Graph Neural Networks in Particle Physics
• QCD or What?
• Cormorant: Covariant Molecular Neural Networks
• Variational Autoencoders for New Physics Mining at the Large Hadron Collider
• The Metric Space of Collider Events
• An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging
• e3nn: Euclidean Neural Networks
• Anomaly detection in high-energy physics using a quantum autoencoder
• Calorimetry with Deep Learning: Particle Simulation and Reconstruction for Collider Physics
• Enhanced Higgs to ${\tau }^{+}{\tau }^{-}$ Searches with Deep Learning
• Anomaly detection with Convolutional Graph Neural Networks
• Autoencoders on FPGAs for real-time, unsupervised new physics detection at 40 MHz at the Large Hadron Collider
• Bump Hunting in Latent Space
• A reconfigurable neural network ASIC for detector front-end data compression at the HL-LHC
• Deep Learning as a Parton Shower
• Quasi Anomalous Knowledge: Searching for new physics with embedded knowledge
• Charged particle tracking via edge-classifying interaction networks
• Autoencoders for Semivisible Jet Detection
• Symmetries, Safety, and Self-Supervision
• Energy-weighted Message Passing: an infra-red and collinear safe graph neural network algorithm
• IRC-safe Graph Autoencoder for unsupervised anomaly detection
• Particle-based Fast Jet Simulation at the LHC with Variational Autoencoders
• The importance of calorimetry for highly-boosted jet substructure
• Matrix Element Regression with Deep Neural Networks -- breaking the CPU barrier

Cited by in corpus (12)

• Anomalies, Representations, and Self-Supervision
• Does Lorentz-symmetric design boost network performance in jet physics?
• Equivariant, Safe and Sensitive -- Graph Networks for New Physics
• Semi-visible jets, energy-based models, and self-supervision
• Hypergraphs in LHC Phenomenology -- The Next Frontier of IRC-Safe Feature Extraction
• Interpretable deep learning models for the inference and classification of LHC data
• ${\mathbb{Z}}_2\times {\mathbb{Z}}_2$ Equivariant Quantum Neural Networks: Benchmarking against Classical Neural Networks
• Semi-supervised permutation invariant particle-level anomaly detection
• Foundations of automatic feature extraction at LHC--point clouds and graphs
• jBOT: Semantic Jet Representation Clustering Emerges from Self-Distillation
• Graph theory inspired anomaly detection at the LHC
• Learnable cut flow for high energy physics