Machine-learning-based prediction of parameters of secondaries in hadronic showers using calorimetric observables

TL;DR

This paper introduces a neural network approach to predict secondary particle parameters in hadronic showers using calorimetric data, aiding in the validation and understanding of complex particle interactions.

Contribution

It presents a novel deep learning method to analyze secondary particle distributions in hadronic showers with highly granular calorimeter data, improving simulation validation.

Findings

Neural network accurately predicts neutron numbers and neutral pion energies.

Model trained on Geant4 simulation data demonstrates effective performance.

Potential for enhanced validation of hadronic shower simulations.

Abstract

The paper describes a novel neural-network-based approach to study the distributions of secondaries produced in hadronic showers using observables provided by highly granular calorimeters. The response is analysed of the highly granular scintillator-steel hadron calorimeter to negative pions with momenta from 10 to 80 GeV simulated with two physics lists from the Geant4 package version 10.3. Several global observables, which characterise different aspects of hadronic shower development, are used as inputs for a deep neural network. The network regression model is trained using a supervised learning and exploiting true information from the simulations. The trained model is applied to predict a number of neutrons and energy of neutral pions produced within a hadronic shower. The achieved performance and possible application of the model to validation of simulations are discussed.