Lorentz Group Equivariant Neural Network for Particle Physics

Alexander Bogatskiy, Brandon Anderson, Jan T. Offermann, Marwah Roussi, David W. Miller +1 more
6/8/2020
hep-phcs.LGhep-ex

Official code: fizisist/LorentzGroupNetwork (49 stars, last commit 2022-03-10).

Abstract

We present a neural network architecture that is fully equivariant with respect to transformations under the Lorentz group, a fundamental symmetry of space and time in physics. The architecture is based on the theory of the finite-dimensional representations of the Lorentz group and the equivariant nonlinearity involves the tensor product. For classification tasks in particle physics, we demonstrate that such an equivariant architecture leads to drastically simpler models that have relatively few learnable parameters and are much more physically interpretable than leading approaches that use CNNs and point cloud approaches. The competitive performance of the network is demonstrated on a public classification dataset [27] for tagging top quark decays given energy-momenta of jet constituents produced in proton-proton collisions.

View on arXivView PDF

Code Implementations(1)

fizisist/LorentzGroupNetworkOfficial100%

Neural network architecture that is fully equivariant with respect to transformations under the Lorentz group, a fundamental symmetry of space and time in physics.

495C, C++Jun 7, 20204 years agoNOASSERTION

Similar Papers

A General Framework for Equivariant Neural Networks on Reductive Lie Groups

87% match2023

Explainable Equivariant Neural Networks for Particle Physics: PELICAN

86% match1 repo142023

Equivariance Is Not All You Need: Characterizing the Utility of Equivariant Graph Neural Networks for Particle Physics Tasks

84% match2023

Symmetry Discovery for Different Data Types

83% match2024

A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups

83% match1 repo2902021

Approximately Equivariant Neural Processes

83% match1 repo2024

Cite this paper

@article{bogatskiy2020lorentz,
  title  = {Lorentz Group Equivariant Neural Network for Particle Physics},
  author = {Alexander Bogatskiy and Brandon Anderson and Jan T. Offermann and Marwah Roussi and David W. Miller and Risi Kondor},
  year   = {2020},
  eprint = {2006.04780},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2006.04780}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2006.04780.svg)](https://codeofpaper.com/paper/2006.04780)

Discussion