Learning to Isolate Muons

TL;DR

This paper demonstrates that advanced neural networks analyzing calorimeter data can better distinguish prompt muons from background than traditional isolation cone methods, revealing new information in calorimeter deposits.

Contribution

It introduces neural network approaches that utilize calorimeter data for muon isolation, surpassing traditional methods and identifying high-level observables that retain discrimination power.

Findings

Neural networks outperform isolation cones in muon classification.

Calorimeter energy distribution contains additional discriminative information.

High-level observables can approximate neural network performance.

Abstract

Distinguishing between prompt muons produced in heavy boson decay and muons produced in association with heavy-flavor jet production is an important task in analysis of collider physics data. We explore whether there is information available in calorimeter deposits that is not captured by the standard approach of isolation cones. We find that convolutional networks and particle-flow networks accessing the calorimeter cells surpass the performance of isolation cones, suggesting that the radial energy distribution and the angular structure of the calorimeter deposits surrounding the muon contain unused discrimination power. We assemble a small set of high-level observables which summarize the calorimeter information and close the performance gap with networks which analyze the calorimeter cells directly. These observables are theoretically well-defined and can be studied with collider data.