Barrier distribution extraction via Gaussian process regression

TL;DR

This paper introduces a Gaussian process regression method for extracting nuclear potential barrier distributions from experimental fusion cross sections, providing a flexible and uncertainty-quantifying alternative to traditional techniques.

Contribution

It presents a novel GPR-based approach for barrier distribution extraction from experimental data, enhancing robustness and uncertainty quantification.

Findings

GPR effectively models complex cross section behaviors.

The method quantifies uncertainties in barrier distributions.

GPR shows robustness to experimental noise.

Abstract

This work presents a novel method for extracting potential barrier distributions from experimental fusion cross sections. We utilize a simple Gaussian process regression (GPR) framework to model the observed cross sections as a function of energy for three nuclear systems. The GPR approach offers a flexible way to represent the experimental data, accommodating potentially complex behavior without introducing strong prior assumptions. This method is applied directly to experimental data and is compared to the traditional direct extraction technique. We discuss the advantages of GPR-based barrier distribution extraction, including the capability to quantify uncertainties and robustness to noise in the experimental data.