Applications of Neural Networks in Hadron Physics

TL;DR

This paper reviews Bayesian neural networks and explores their application in hadron physics, specifically analyzing two-photon exchange effects and cross section ratios with uncertainty estimations.

Contribution

It introduces the use of Bayesian neural networks in hadron physics, focusing on model comparison, uncertainty estimation, and over-fitting issues in the context of two-photon exchange studies.

Findings

Predictions of cross section ratios with uncertainty estimates.

Comparison of different models for systematic uncertainty.

Insights into over-fitting and model selection in neural network applications.

Abstract

The Bayesian approach for the feed-forward neural networks is reviewed. Its potential for usage in hadron physics is discussed. As an example of the application the study of the the two-photon exchange effect is presented. We focus on the model comparison, the estimation of the systematic uncertainties due to the choice of the model, and the over-fitting. As an illustration the predictions of the cross sections ratio $d \sigma(e^+ p\to e^+ p)/d \sigma(e^- p\to e^- p)$ are given together with the estimate of the uncertainty due to the parametrization choice.