Nuclear Data Evaluation Utilizing Bayesian Monte Carlo Methods
Abstract
While the body of experimental nuclear data continues to grow at a rapid pace, significant gaps persist. The lack of evaluated experimental data is far more pronounced in charged-particle reactions which are essential to a large portion of the isotope production community for vital medical, industrial, and national security interests. In the absence of experimental and evaluated data, isotope production relies on theoretical models that may result in low yield and isotopic impurities. This work investigates model and parameter adjustments in TALYS 2.0 to establish uncertainty bands on proton-induced cross sections, highlight influential parameters in calculations, and constrain adjustments within physically defensible bounds. The EXFOR dataset for proton-induced reactions on natural copper, comprised of 24 independent cross sections, was used for iterative Bayesian model updating. Upon convergence, parameters and their uncertainties were compared against 23 cumulative cross sections from EXFOR, highlighting the limitations of generalization. Various machine learning models were leveraged to predict TALYS performance, with a small neural network showing useful ranking capabilities across several generations, suggesting the use of a surrogate approach to screening for future evaluations.
Related
- Nuclear Mass Models with Deformation Optimized and Diagnosed by Bayesian Analysis with Markov Chain Monte Carlo
- Wigner distribution modeling of nuclear state in NuWro Monte Carlo event generator
- Global Analysis of Nuclear Parton Distribution Functions using Markov Chain Monte Carlo
- A Lightweight Python-based Monte Carlo Framework for Visualizing Energy-dependent Dose Calibrator Response: A Nuclear Medicine Education Tool
- Time-Dependent Quantum Monte Carlo for fermions: from Bayesian conditioning to spinor dynamics
BibTeX
@misc{apgar2026nuclear,
title = {{Nuclear Data Evaluation Utilizing Bayesian Monte Carlo Methods}},
author = {Catherine E. Apgar and Lee A. Bernstein and Yun-Hsuan Lee and Wilson Lin and Alan B. McIntosh and Dmitri G. Medvedev and Jonathan T. Morrell and Ellen M. O'Brien and C. Etienne Vermeulen and Andrew S. Voyles},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.36338},
eprint = {2610.36338},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.36338},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Nuclear Data Evaluation Utilizing Bayesian Monte Carlo Methods
AU - Catherine E. Apgar
AU - Lee A. Bernstein
AU - Yun-Hsuan Lee
AU - Wilson Lin
AU - Alan B. McIntosh
AU - Dmitri G. Medvedev
AU - Jonathan T. Morrell
AU - Ellen M. O'Brien
AU - C. Etienne Vermeulen
AU - Andrew S. Voyles
PY - 2026
DA - 2026/10/26
PB - arXiv
DO - 10.48550/arxiv.2610.36338
UR - https://arxiv.org/abs/2610.36338
AB - While the body of experimental nuclear data continues to grow at a rapid pace, significant gaps persist. The lack of evaluated experimental data is far more pronounced in charged-particle reactions which are essential to a large portion of the isotope production community for vital medical, industrial, and national security interests. In the absence of experimental and evaluated data, isotope production relies on theoretical models that may result in low yield and isotopic impurities. This work investigates model and parameter adjustments in TALYS 2.0 to establish uncertainty bands on proton-induced cross sections, highlight influential parameters in calculations, and constrain adjustments within physically defensible bounds. The EXFOR dataset for proton-induced reactions on natural copper, comprised of 24 independent cross sections, was used for iterative Bayesian model updating. Upon convergence, parameters and their uncertainties were compared against 23 cumulative cross sections from EXFOR, highlighting the limitations of generalization. Various machine learning models were leveraged to predict TALYS performance, with a small neural network showing useful ranking capabilities across several generations, suggesting the use of a surrogate approach to screening for future evaluations.
ER -