Estimating event-by-event multiplicity by a Machine Learning Method for Hadronization Studies

TL;DR

This paper demonstrates that deep learning, specifically ResNet neural networks, can effectively predict event-by-event charged particle multiplicities in proton-proton collisions, offering a new approach to studying non-perturbative hadronization processes.

Contribution

The study introduces the application of ResNet neural networks to predict charged particle multiplicities, providing a novel machine learning approach to hadronization modeling.

Findings

Neural networks with over 1000 parameters can predict multiplicities up to N_ch ≈ 90.

The method achieves predictions based on training with Lund string fragmentation model data.

Deep learning models can capture non-linear features of hadronization processes.

Abstract

Hadronization is a non-perturbative process, which theoretical description can not be deduced from first principles. Modeling hadron formation requires several assumptions and various phenomenological approaches. Utilizing state-of-the-art Deep Learning algorithms, it is eventually possible to train neural networks to learn non-linear and non-perturbative features of the physical processes. In this study, the prediction results of three trained ResNet networks are presented, by investigating charged particle multiplicities at event-by-event level. The widely used Lund string fragmentation model is applied as a training-baseline at $\sqrt{s}= 7$ TeV proton-proton collisions. We found that neural-networks with $ \gtrsim\mathcal{O}(10^3)$ parameters can predict the event-by-event charged hadron multiplicity values up to $ N_\mathrm{ch}\lesssim 90 $.