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

Title: Deep learning: Extrapolation tool for ab initio nuclear theory

Abstract

Ab initio approaches in nuclear theory, such as the no-core shell model (NCSM), have been developed for approximately solving finite nuclei with realistic strong interactions. The NCSM and other approaches require an extrapolation of the results obtained in a finite basis space to the infinite basis space limit and assessment of the uncertainty of those extrapolations. Each observable requires a separate extrapolation and many observables have no proven extrapolation method. We propose a feed-forward artificial neural network (ANN) method as an extrapolation tool to obtain the ground-state energy and the ground-state point-proton root-mean-square (rms) radius along with their extrapolation uncertainties. The designed ANNs are sufficient to produce results for these two very different observables in 6Li from the ab initio NCSM results in small basis spaces that satisfy the following theoretical physics condition: independence of basis space parameters in the limit of extremely large matrices. Comparisons of the ANN results with other extrapolation methods are also provided.

Authors:
 [1];  [2];  [3];  [2];  [4];  [5];  [5];  [6];  [6];  [2];  [7]
  1. Iowa State Univ., Ames, IA (United States). Dept. of Computer Science; Horia Hulubei National Inst. for Physics and Nuclear Engineering (Romania)
  2. Iowa State Univ., Ames, IA (United States). Dept. of Physics and Astronomy
  3. Iowa State Univ., Ames, IA (United States). Dept. of Mathematics
  4. Moscow State Univ., Moscow (Russian Federation). Skobeltsyn Inst. of Nuclear Physics; Pacific National Univ., Khabarovsk (Russia). Dept. of Physics
  5. Inst. for Basic Science, Daejeon (Korea). Rare Isotope Science Project
  6. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
  7. Iowa State Univ., Ames, IA (United States). Dept. of Computer Science
Publication Date:
Research Org.:
Lawrence Berkeley National Laboratory-National Energy Research Scientific Computing Center (NERSC)
Sponsoring Org.:
USDOE Office of Science (SC)
OSTI Identifier:
1527179
DOE Contract Number:  
FG02-87ER40371; SC000018223; AC02-05CH11231
Resource Type:
Journal Article
Journal Name:
Physical Review C
Additional Journal Information:
Journal Volume: 99; Journal Issue: 5; Journal ID: ISSN 2469-9985
Publisher:
American Physical Society (APS)
Country of Publication:
United States
Language:
English

Citation Formats

Negoita, Gianina Alina, Vary, James P., Luecke, Glenn R., Maris, Pieter, Shirokov, Andrey M., Shin, Ik Jae, Kim, Youngman, Ng, Esmond G., Yang, Chao, Lockner, Matthew, and Prabhu, Gurpur M. Deep learning: Extrapolation tool for ab initio nuclear theory. United States: N. p., 2019. Web. doi:10.1103/PhysRevC.99.054308.
Negoita, Gianina Alina, Vary, James P., Luecke, Glenn R., Maris, Pieter, Shirokov, Andrey M., Shin, Ik Jae, Kim, Youngman, Ng, Esmond G., Yang, Chao, Lockner, Matthew, & Prabhu, Gurpur M. Deep learning: Extrapolation tool for ab initio nuclear theory. United States. doi:10.1103/PhysRevC.99.054308.
Negoita, Gianina Alina, Vary, James P., Luecke, Glenn R., Maris, Pieter, Shirokov, Andrey M., Shin, Ik Jae, Kim, Youngman, Ng, Esmond G., Yang, Chao, Lockner, Matthew, and Prabhu, Gurpur M. Wed . "Deep learning: Extrapolation tool for ab initio nuclear theory". United States. doi:10.1103/PhysRevC.99.054308.
@article{osti_1527179,
title = {Deep learning: Extrapolation tool for ab initio nuclear theory},
author = {Negoita, Gianina Alina and Vary, James P. and Luecke, Glenn R. and Maris, Pieter and Shirokov, Andrey M. and Shin, Ik Jae and Kim, Youngman and Ng, Esmond G. and Yang, Chao and Lockner, Matthew and Prabhu, Gurpur M.},
abstractNote = {Ab initio approaches in nuclear theory, such as the no-core shell model (NCSM), have been developed for approximately solving finite nuclei with realistic strong interactions. The NCSM and other approaches require an extrapolation of the results obtained in a finite basis space to the infinite basis space limit and assessment of the uncertainty of those extrapolations. Each observable requires a separate extrapolation and many observables have no proven extrapolation method. We propose a feed-forward artificial neural network (ANN) method as an extrapolation tool to obtain the ground-state energy and the ground-state point-proton root-mean-square (rms) radius along with their extrapolation uncertainties. The designed ANNs are sufficient to produce results for these two very different observables in 6Li from the ab initio NCSM results in small basis spaces that satisfy the following theoretical physics condition: independence of basis space parameters in the limit of extremely large matrices. Comparisons of the ANN results with other extrapolation methods are also provided.},
doi = {10.1103/PhysRevC.99.054308},
journal = {Physical Review C},
issn = {2469-9985},
number = 5,
volume = 99,
place = {United States},
year = {2019},
month = {5}
}

Works referenced in this record:

Nuclear mass predictions for the crustal composition of neutron stars: A Bayesian neural network approach
journal, January 2016


Large-scale exact diagonalizations reveal low-momentum scales of nuclei
journal, March 2018


Ab initio no core shell model
journal, March 2013

  • Barrett, Bruce R.; Navrátil, Petr; Vary, James P.
  • Progress in Particle and Nuclear Physics, Vol. 69
  • DOI: 10.1016/j.ppnp.2012.10.003

Nucleon-nucleon interaction in the J -matrix inverse scattering approach and few-nucleon systems
journal, October 2004


Consistent empirical physical formulas for potential energy curves of 38–66Ti isotopes by using neural networks
journal, November 2013

  • Akkoyun, S.; Bayram, T.; Kara, S. O.
  • Physics of Particles and Nuclei Letters, Vol. 10, Issue 6
  • DOI: 10.1134/S1547477113060022

Spurious center-of-mass motion
journal, December 1974


Neural networks for impact parameter determination
journal, May 1996


Scaling of ab-initio nuclear physics calculations on multicore computer architectures
journal, May 2010


Energy levels of light nuclei A=5, 6, 7
journal, September 2002


Infrared length scale and extrapolations for the no-core shell model
journal, June 2015


Approximation by superpositions of a sigmoidal function
journal, December 1989

  • Cybenko, G.
  • Mathematics of Control, Signals, and Systems, Vol. 2, Issue 4
  • DOI: 10.1007/BF02551274

Coulomb-Sturmian basis for the nuclear many-body problem
journal, September 2012


Accurate charge-dependent nucleon-nucleon potential at fourth order of chiral perturbation theory
journal, October 2003


Center-of-Mass Motion in Brueckner Theory for a Finite Nucleus
journal, March 1958


Accurate nucleon–nucleon potential based upon chiral perturbation theory
journal, January 2002


Ab initio nuclear structure – the large sparse matrix eigenvalue problem
journal, July 2009


Loosely bound three-body nuclear systems in the J-matrix approach
journal, August 2004


Impact parameter determination for heavy-ion collisions by use of a neural network
journal, March 1995


AB INITIO NUCLEAR STRUCTURE CALCULATIONS OF p-SHELL NUCLEI WITH JISP16
journal, July 2013


Training feedforward networks with the Marquardt algorithm
journal, January 1994

  • Hagan, M. T.; Menhaj, M. B.
  • IEEE Transactions on Neural Networks, Vol. 5, Issue 6
  • DOI: 10.1109/72.329697

An Algorithm for Least-Squares Estimation of Nonlinear Parameters
journal, June 1963

  • Marquardt, Donald W.
  • Journal of the Society for Industrial and Applied Mathematics, Vol. 11, Issue 2
  • DOI: 10.1137/0111030

Recent experimental progress in nuclear halo structure studies
journal, January 2013

  • Tanihata, Isao; Savajols, Herve; Kanungo, Rituparna
  • Progress in Particle and Nuclear Physics, Vol. 68
  • DOI: 10.1016/j.ppnp.2012.07.001

Refining mass formulas for astrophysical applications: A Bayesian neural network approach
journal, October 2017


Multilayer feedforward networks are universal approximators
journal, January 1989


N3LO NN interaction adjusted to light nuclei in ab exitu approach
journal, October 2016


Similarity renormalization group for nucleon-nucleon interactions
journal, June 2007


Bayesian Interpolation
journal, May 1992


Improving the scalability of a symmetric iterative eigensolver for multi-core platforms: IMPROVING THE SCALABILITY OF A SYMMETRIC ITERATIVE EIGENSOLVER
journal, September 2013

  • Aktulga, Hasan Metin; Yang, Chao; Ng, Esmond G.
  • Concurrency and Computation: Practice and Experience, Vol. 26, Issue 16
  • DOI: 10.1002/cpe.3129

Halo nuclei He 6 and He 8 with the Coulomb-Sturmian basis
journal, September 2014


Universal properties of infrared oscillator basis extrapolations
journal, April 2013


Ab initio no-core solutions for 6 Li
journal, May 2017

  • Shin, Ik Jae; Kim, Youngman; Maris, Pieter
  • Journal of Physics G: Nuclear and Particle Physics, Vol. 44, Issue 7
  • DOI: 10.1088/1361-6471/aa6cb7

Nuclear charge radii: density functional theory meets Bayesian neural networks
journal, October 2016

  • Utama, R.; Chen, Wei-Chia; Piekarewicz, J.
  • Journal of Physics G: Nuclear and Particle Physics, Vol. 43, Issue 11
  • DOI: 10.1088/0954-3899/43/11/114002

An artificial neural network application on nuclear charge radii
journal, March 2013


Covariant spectator theory of np scattering: Phase shifts obtained from precision fits to data below 350 MeV
journal, July 2008


Decoding β -decay systematics: A global statistical model for β half-lives
journal, October 2009


Validating neural-network refinements of nuclear mass models
journal, January 2018


Bayesian approach to model-based extrapolation of nuclear observables
journal, September 2018


Convergence properties of ab initio calculations of light nuclei in a harmonic oscillator basis
journal, November 2012


Corrections to nuclear energies and radii in finite oscillator spaces
journal, September 2012


Corrections to nucleon capture cross sections computed in truncated Hilbert spaces
journal, March 2017


Ab initio no-core full configuration calculations of light nuclei
journal, January 2009


Nuclear mass systematics using neural networks
journal, November 2004


Natural orbital description of the halo nucleus 6He
journal, November 2017

  • Constantinou, Chrysovalantis; Caprio, Mark A.; Vary, James P.
  • Nuclear Science and Techniques, Vol. 28, Issue 12
  • DOI: 10.1007/s41365-017-0332-6

Impact parameter determination in experimental analysis using a neural network
journal, March 1997


Infrared extrapolations of quadrupole moments and transitions
journal, April 2016


From low-momentum interactions to nuclear structure
journal, July 2010

  • Bogner, S. K.; Furnstahl, R. J.; Schwenk, A.
  • Progress in Particle and Nuclear Physics, Vol. 65, Issue 1
  • DOI: 10.1016/j.ppnp.2010.03.001