skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Bayesian approach to model-based extrapolation of nuclear observables

Abstract

Here, we considered 10 global models based on nuclear density functional theory with realistic energy density functionals as well as two more phenomenological mass models. The emulators of two-neutron separation energy residuals and Bayesian confidence intervals defining theoretical error bars were constructed using Bayesian Gaussian processes and Bayesian neural networks. By establishing statistical methodology and parameters, we carried out extrapolations toward the two-neutron dripline. While both Gaussian processes (GP) and Bayesian neural networks reduce the root-mean-square (rms) deviation from experiment significantly, GP offers a better and much more stable performance. The increase in the predictive power of microscopic models aided by the statistical treatment is quite astonishing: The resulting rms deviations from experiment on the testing dataset are similar to those of more phenomenological models. The estimated credibility intervals on predictions make it possible to evaluate predictive power of individual models and also make quantified predictions using groups of models. The proposed robust statistical approach to extrapolation of nuclear model results can be useful for assessing the impact of current and future experiments in the context of model developments. The new Bayesian capability to evaluate residuals is also expected to impact research in the domains where experiments are currently impossible,more » for instance, in simulations of the astrophysical r process.« less

Authors:
 [1];  [1];  [1];  [1]
  1. Michigan State Univ., East Lansing, MI (United States)
Publication Date:
Research Org.:
Michigan State Univ., East Lansing, MI (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1491246
Alternate Identifier(s):
OSTI ID: 1472205
Grant/Contract Number:  
NA0002847; NA0002574; SC0018083; SC0013365
Resource Type:
Accepted Manuscript
Journal Name:
Physical Review C
Additional Journal Information:
Journal Volume: 98; Journal Issue: 3; Journal ID: ISSN 2469-9985
Publisher:
American Physical Society (APS)
Country of Publication:
United States
Language:
English
Subject:
73 NUCLEAR PHYSICS AND RADIATION PHYSICS

Citation Formats

Neufcourt, Léo, Cao, Yuchen, Nazarewicz, Witold, and Viens, Frederi. Bayesian approach to model-based extrapolation of nuclear observables. United States: N. p., 2018. Web. doi:10.1103/PhysRevC.98.034318.
Neufcourt, Léo, Cao, Yuchen, Nazarewicz, Witold, & Viens, Frederi. Bayesian approach to model-based extrapolation of nuclear observables. United States. doi:10.1103/PhysRevC.98.034318.
Neufcourt, Léo, Cao, Yuchen, Nazarewicz, Witold, and Viens, Frederi. Mon . "Bayesian approach to model-based extrapolation of nuclear observables". United States. doi:10.1103/PhysRevC.98.034318. https://www.osti.gov/servlets/purl/1491246.
@article{osti_1491246,
title = {Bayesian approach to model-based extrapolation of nuclear observables},
author = {Neufcourt, Léo and Cao, Yuchen and Nazarewicz, Witold and Viens, Frederi},
abstractNote = {Here, we considered 10 global models based on nuclear density functional theory with realistic energy density functionals as well as two more phenomenological mass models. The emulators of two-neutron separation energy residuals and Bayesian confidence intervals defining theoretical error bars were constructed using Bayesian Gaussian processes and Bayesian neural networks. By establishing statistical methodology and parameters, we carried out extrapolations toward the two-neutron dripline. While both Gaussian processes (GP) and Bayesian neural networks reduce the root-mean-square (rms) deviation from experiment significantly, GP offers a better and much more stable performance. The increase in the predictive power of microscopic models aided by the statistical treatment is quite astonishing: The resulting rms deviations from experiment on the testing dataset are similar to those of more phenomenological models. The estimated credibility intervals on predictions make it possible to evaluate predictive power of individual models and also make quantified predictions using groups of models. The proposed robust statistical approach to extrapolation of nuclear model results can be useful for assessing the impact of current and future experiments in the context of model developments. The new Bayesian capability to evaluate residuals is also expected to impact research in the domains where experiments are currently impossible, for instance, in simulations of the astrophysical r process.},
doi = {10.1103/PhysRevC.98.034318},
journal = {Physical Review C},
number = 3,
volume = 98,
place = {United States},
year = {2018},
month = {9}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 15 works
Citation information provided by
Web of Science

Save / Share: