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

Journal Article · · Physical Review C

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.

Research Organization:
Michigan State Univ., East Lansing, MI (United States)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA)
Grant/Contract Number:
NA0002847; NA0002574; SC0018083; SC0013365
OSTI ID:
1491246
Alternate ID(s):
OSTI ID: 1472205
Journal Information:
Physical Review C, Vol. 98, Issue 3; ISSN 2469-9985
Publisher:
American Physical Society (APS)Copyright Statement
Country of Publication:
United States
Language:
English
Citation Metrics:
Cited by: 117 works
Citation information provided by
Web of Science

References (90)

Nuclear mass predictions for the crustal composition of neutron stars: A Bayesian neural network approach journal January 2016
New relativistic mean-field interaction with density-dependent meson-nucleon couplings journal February 2005
Rates of convergence of the Hastings and Metropolis algorithms journal February 1996
Uncertainties of mass extrapolations in Hartree-Fock-Bogoliubov mass models journal May 2014
Microscopic mass formulas journal July 1995
Probabilistic forecasts, calibration and sharpness journal April 2007
Propagation of uncertainties in the nuclear DFT models journal February 2015
The limits of the nuclear landscape journal June 2012
Nuclear energy density optimization: Shell structure journal May 2014
Hartree-Fock-Bogolyubov description of nuclei near the neutron-drip line journal June 1984
Predictive power of nuclear-mass models journal July 2014
Performance of the Levenberg–Marquardt neural network approach in nuclear mass prediction journal March 2017
Relativistic mean-field interaction with density-dependent meson-nucleon vertices based on microscopical calculations journal November 2011
David Draper and E. I. George, and a rejoinder by the authors journal November 1999
The Ame2003 atomic mass evaluation journal December 2003
The AME2016 atomic mass evaluation (II). Tables, graphs and references journal March 2017
The Ame2003 atomic mass evaluation journal December 2003
The AME2016 atomic mass evaluation (I). Evaluation of input data; and adjustment procedures journal March 2017
A study on ground-state energies of nuclei by using neural networks journal January 2014
Nuclear ground-state masses and deformations: FRDM(2012) journal May 2016
Precision Mass Measurements on Neutron-Rich Rare-Earth Isotopes at JYFLTRAP: Reduced Neutron Pairing and Implications for r -Process Calculations journal June 2018
Variations on a theme by Skyrme: A systematic study of adjustments of model parameters journal March 2009
The effective force NL3 revisited journal January 2009
Strictly Proper Scoring Rules, Prediction, and Estimation journal March 2007
Estimating Parameter Uncertainty in Binding-Energy Models by the Frequency-Domain Bootstrap journal December 2017
Global performance of covariant energy density functionals: Ground state observables of even-even nuclei and the estimate of theoretical uncertainties journal May 2014
Towards a better parametrisation of Skyrme-like effective forces: A critical study of the SkM force journal September 1982
Neutron drip line: Single-particle degrees of freedom and pairing properties as sources of theoretical uncertainties journal January 2015
Uncertainty decomposition method and its application to the liquid drop model journal March 2016
New Skyrme effective forces for supernovae and neutron rich nuclei journal January 1995
Structure of even-even nuclei using a mapped collective Hamiltonian and the D1S Gogny interaction journal January 2010
Positioning the neutron drip line and the r-process paths in the nuclear landscape journal September 2015
Uncertainty propagation within the UNEDF models journal March 2017
Global study of quadrupole correlation effects journal March 2006
Valence p - n interactions and the development of collectivity in heavy nuclei journal February 1987
Error estimates of theoretical models: a guide journal May 2014
Nuclear mass predictions based on Bayesian neural network approach with pairing and shell effects journal March 2018
Nuclear energy density optimization: Large deformations journal February 2012
Uncertainty quantification and propagation in nuclear density functional theory journal December 2015
Propagation of uncertainties in the Skyrme energy-density-functional model journal March 2013
Refining mass formulas for astrophysical applications: A Bayesian neural network approach journal October 2017
Addressing spectroscopic quality of covariant density functional theory journal February 2015
Covariant energy density functionals: Nuclear matter constraints and global ground state properties journal May 2016
Nuclidic Mass Formula on a Spherical Basis with an Improved Even-Odd Term journal February 2005
On the Markov chain central limit theorem journal January 2004
Bayesian Learning for Neural Networks book January 1996
Bayesian calibration of computer models journal August 2001
Nuclear energy density optimization journal August 2010
Nuclear charge radii: density functional theory meets Bayesian neural networks journal October 2016
The limits of the nuclear landscape explored by the relativistic continuum Hartree–Bogoliubov theory journal May 2018
Self-consistent mean-field models for nuclear structure journal January 2003
One-quasiparticle states in the nuclear energy density functional theory journal February 2010
Markov Chain Monte Carlo in Practice book December 1995
Validating neural-network refinements of nuclear mass models journal January 2018
Relativistic nuclear energy density functionals: Adjusting parameters to binding energies journal September 2008
Further explorations of Skyrme-Hartree-Fock-Bogoliubov mass formulas. XVI. Inclusion of self-energy effects in pairing journal March 2016
Fluctuating parts of nuclear ground-state correlation energies journal May 2013
The Ame2012 atomic mass evaluation journal December 2012
Bayesian Model Selection and Model Averaging journal March 2000
Nuclear mass systematics using neural networks journal November 2004
Further explorations of Skyrme-Hartree-Fock-Bogoliubov mass formulas. XIII. The 2012 atomic mass evaluation and the symmetry coefficient journal August 2013
Global microscopic calculations of ground-state spins and parities for odd-mass nuclei journal August 2007
Optimizing relativistic energy density functionals: covariance analysis journal February 2015
An approach to adjustment of relativistic mean field model parameters journal January 2017
Uncertainty Quantification for Nuclear Density Functional Theory and Information Content of New Measurements journal March 2015
Gamow-Hartree-Fock-Bogoliubov method: Representation of quasiparticles with Berggren sets of wave functions journal October 2008
The Ame2012 atomic mass evaluation journal December 2012
Towards a better parametrisation of Skyrme-like effective forces: A critical study of the SkM force text January 1982
Markov Chain Monte Carlo in Practice journal August 1997
Gamow-Hartree-Fock-Bogoliubov Method: Representation of quasiparticles with Berggren sets of wave functions text January 2008
Relativistic Nuclear Energy Density Functionals: adjusting parameters to binding energies text January 2008
The effective force NL3 revisited text January 2009
One-quasiparticle States in the Nuclear Energy Density Functional Theory text January 2009
Structure of even-even nuclei using a mapped collective Hamiltonian and the D1S Gogny interaction text January 2009
Fluctuating parts of nuclear ground state correlation energies text January 2012
Error Estimates of Theoretical Models: a Guide text January 2014
Global performance of covariant energy density functionals: ground state observables of even-even nuclei and the estimate of theoretical uncertainties text January 2014
Positioning the neutron drip line and the r-process paths in the nuclear landscape text January 2014
Uncertainty Quantification for Nuclear Density Functional Theory and Information Content of New Measurements text January 2015
Covariant energy density functionals: nuclear matter constraints and global ground state properties text January 2016
Nuclear charge radii: Density functional theory meets Bayesian neural networks text January 2016
Uncertainty propagation within the UNEDF models text January 2016
Refining mass formulas for astrophysical applications: a Bayesian neural network approach text January 2017
The limits of the nuclear landscape explored by the relativistic continuum Hatree-Bogoliubov theory text January 2017
Validating neural-network refinements of nuclear mass models text January 2017
Nuclear mass predictions based on Bayesian neural network approach with pairing and shell effects text January 2018
Precision mass measurements on neutron-rich rare-earth isotopes at JYFLTRAP - reduced neutron pairing and implications for the $r$-process calculations text January 2018
Global study of quadrupole correlation effects text January 2005
Global microscopic calculations of ground-state spin and parity for odd-mass nuclei text January 2007
Microscopic mass formulae text January 1994

Cited By (9)

Alpha half-lives calculation of superheavy nuclei with Q α -value predictions based on the Bayesian neural network approach journal October 2019
Influence of nuclear mass uncertainties on radiative neutron-capture rates journal August 2019
A Bayesian analysis of longitudinal farm surveys in Central Malawi reveals yield determinants and site-specific management strategies journal August 2019
Evaluating Magnetocaloric Effect in Magnetocaloric Materials: A Novel Approach Based on Indirect Measurements Using Artificial Neural Networks journal May 2019
Deep learning: Extrapolation tool for ab initio nuclear theory text January 2018
Neutron drip line in the Ca region from Bayesian model averaging text January 2019
Electroweak probes of ground state densities text January 2019
Beyond the proton drip line: Bayesian analysis of proton-emitting nuclei text January 2019
Impact of statistical uncertainties on the composition of the outer crust of a neutron star text January 2019

Figures / Tables (14)


Similar Records