Bayesian Learning of (n,p) Reaction Cross Sections with Quantified Uncertainties

TL;DR

This paper introduces a Bayesian Neural Network framework to predict neutron-induced (n,p) reaction cross sections with quantified uncertainties, improving data-driven nuclear data evaluations especially where experimental data are limited.

Contribution

The work presents a novel BNN-based method incorporating physical features and uncertainty quantification for (n,p) cross section prediction, benchmarked against existing data and models.

Findings

BNN-I6 achieves accurate cross section predictions with reliable uncertainty estimates.

The model outperforms or matches existing data libraries like TENDL-2023.

Feature importance analysis highlights the influence of theoretical inputs.

Abstract

Accurate modeling of neutron-induced (n,p) reaction cross sections is essential for diverse applications in nuclear physics, including reactor design, nuclear astrophysics, and radionuclide production. However, experimental data are often sparse or incomplete, and theoretical results like TALYS Evaluated Nuclear Data Library (TENDL-2023) data may carry systematic uncertainties. In this work, we present a data-driven framework based on a Bayesian Neural Network (BNN), denoted as BNN-I6, to predict (n,p) reaction cross sections with quantified uncertainties. The model incorporates six physically motivated input features and is trained on Evaluated Nuclear Data from the ENDF/B-VIII.1 library. Leveraging stochastic variational inference, the BNN offers reliable uncertainty estimates in addition to accurate predictions. The performance of BNN-I6 is benchmarked against the TENDL-2023 library and experimental measurements across a wide range of nuclei. Additionally, SHapley Additive exPlanations (SHAP) based feature-importance analysis reveals the dominant role of theoretical cross-section inputs in driving predictions. These results highlight the potential of BNN-based approaches to enhance nuclear data evaluations and support future applications in data-scarce regimes.