A fast Bayesian surrogate for the photon flux in ultra-peripheral collisions

TL;DR

This paper introduces a Bayesian neural network-based surrogate model that significantly accelerates the computation of photon flux in ultraperipheral collisions, enabling rapid uncertainty propagation and analysis.

Contribution

It presents a novel, fast surrogate for the EPA flux using Bayesian neural networks, reducing computation time by two orders of magnitude and allowing flexible uncertainty analysis.

Findings

Achieved two orders of magnitude faster flux calculations.

Successfully propagated experimental uncertainties on nuclear parameters.

Provided a modular framework for future UPC analyses.

Abstract

We present a fast surrogate for the Equivalent Photon Approximation (EPA) flux in ultraperipheral collisions (UPCs), based on a Bayesian neural network (BNN) trained over analytical flux integrals with an iterative procedure focused on regions of high relative uncertainties to minimise the number of integrations. The surrogate propagates experimentally available uncertainties on the neutron skin thickness and surface diffuseness. Once trained, this surrogate technique brings an estimated gain of two orders of magnitude in CPU time. The implementation provides a modular framework for rapidly propagating updated nuclear priors and assessing uncertainties for photon flux in future UPC analyses.