Exploring the Possibility of a Recovery of Physics Process Properties from a Neural Network Model

TL;DR

This paper investigates whether neural networks trained on simulated QCD data can recover underlying physical process properties, aiming to enhance interpretability and understanding of particle physics mechanisms.

Contribution

The study introduces a simple physical generator and demonstrates that neural networks can recover key process distributions with partial prior knowledge, improving interpretability.

Findings

Neural networks can approximate probability distributions of physical processes.

Partial knowledge of the generator improves interpretability.

Potential application to real QCD data analysis.

Abstract

The application of machine learning methods to particle physics often doesn't provide enough understanding of the underlying physics. An interpretable model which provides a way to improve our knowledge of the mechanism governing a physical system directly from the data can be very useful. In this paper, we introduce a simple artificial physical generator based on the Quantum chromodynamical (QCD) fragmentation process. The data simulated from the generator are then passed to a neural network model which we base only on the partial knowledge of the generator. We aim to see if the interpretation of the generated data can provide the probability distributions of basic processes of such a physical system. This way, some of the information we omitted from the network model on purpose is recovered. We believe this approach can be beneficial in the analysis of real QCD processes.