Machine Learning based KNO-scaling of charged hadron multiplicities with Hijing++

Abstract

The scaling properties of the final state charged hadron and mean jet multiplicity distributions, calculated by deep residual neural network architectures with different complexities are presented. The parton-level input of the neural networks are generated by the Hijing++ Monte Carlo event generator. Hadronization neural networks, trained with $\sqrt{s}=7$ TeV events are utilized to perform predictions for various LHC energies from $\sqrt{s}=0.9$ TeV to 13 TeV. KNO-scaling properties were adopted by the networks at hadronic level.