This website requires JavaScript.

PELICAN: Permutation Equivariant and Lorentz Invariant or Covariant Aggregator Network for Particle Physics

Alexander BogatskiyTimothy HoffmanDavid W. MillerJan T. Offermann
Nov 2022
摘要
Many current approaches to machine learning in particle physics use genericarchitectures that require large numbers of parameters and disregard underlyingphysics principles, limiting their applicability as scientific modeling tools.In this work, we present a machine learning architecture that uses a set ofinputs maximally reduced with respect to the full 6-dimensional Lorentzsymmetry, and is fully permutation-equivariant throughout. We study theapplication of this network architecture to the standard task of top quarktagging and show that the resulting network outperforms all existingcompetitors despite much lower model complexity. In addition, we present aLorentz-covariant variant of the same network applied to a 4-momentumregression task.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答