Taming nuclear complexity with a committee of multilayer neural networks

Abstract

We demonstrate that a committee of deep neural networks is capable of predicting the ground-state and excited energies of more than 1800 atomic nuclei with an accuracy akin to the one achieved by state-of-the-art nuclear energy density functionals (EDFs) and a major speed-up. An active learning strategy is proposed to train this algorithm with a minimal set of 210 nuclei. This approach enables future fast studies of the influence of EDFs parametrizations on structure properties over the whole nuclear chart and suggests that for the first time an artificial intelligence successfully encoded the laws of nuclear deformation.

References in corpus (16)

• Nuclear magic numbers: new features far from stability
• Relativistic Nuclear Energy Density Functionals: Mean-Field and Beyond
• Structure of even-even nuclei using a mapped collective Hamiltonian and the D1S Gogny interaction
• Skyrme-Hartree-Fock-Bogoliubov nuclear mass formulas: Crossing the 0.6 MeV threshold with microscopically deduced pairing
• Nuclear mass predictions based on Bayesian neural network approach with pairing and shell effects
• Bayesian approach to model-based extrapolation of nuclear observables
• Modification of mass formula by considering isospin effects
• Further explorations of Skyrme-Hartree-Fock-Bogoliubov mass formulas. IX: Constraint of pairing force to ${}^1{S}_0$ neutron-matter gap
• Nuclear charge radii: Density functional theory meets Bayesian neural networks
• Refining mass formulas for astrophysical applications: a Bayesian neural network approach
• Beyond relativistic mean-field studies of low-lying states in neutron-deficient krypton isotopes
• An artificial neural network application on nuclear charge radii
• State-of-the-art of beyond mean field theories with nuclear density functionals
• Validating neural-network refinements of nuclear mass models
• FELIX-2.0: New version of the finite element solver for the time dependent generator coordinate method with the Gaussian overlap approximation
• Active Learning Algorithm for Computational Physics

Cited by in corpus (43)

• Machine Learning in Nuclear Physics
• Future of Nuclear Fission Theory
• Nuclear mass predictions with machine learning reaching the accuracy required by $r$-process studies
• Study of Charge Radii with Neural Networks
• Precision Studies of QCD in the Low Energy Domain of the EIC
• Nuclear masses learned from a probabilistic neural network
• Multi-task learning on nuclear masses and separation energies with the kernel ridge regression
• Machine learning the deuteron
• Skyrme-Hartree-Fock-Bogoliubov mass models on a 3D Mesh: Effect of triaxial shape
• Nuclear energy density functionals from machine learning
• Bayesian evaluation of charge yields of fission fragments of 239U
• Skyrme-Hartree-Fock-Bogoliubov mass models on a 3D mesh: II. Time-reversal symmetry breaking
• Training and Projecting: A Reduced Basis Method Emulator for Many-Body Physics
• Nuclear binding energy predictions using neural networks: Application of the multilayer perceptron
• Prediction of Nuclear Charge Density Distribution with Feedback Neural Network
• Recent advances in the quantification of uncertainties in reaction theory
• Application of artificial intelligence in the determination of impact parameter in heavy-ion collisions at intermediate energies
• Optimizing multilayer Bayesian neural networks for evaluation of fission yields
• Extrapolating from neural network models: a cautionary tale
• Taming nucleon density distributions with deep neural network method
• Nuclear mass predictions based on convolutional neural network
• Calibration of nuclear charge density distribution by back-propagation neural networks
• Application of multilayer perceptron with data augmentation in nuclear physics
• Neural Network Emulation of Spontaneous Fission
• The Neutron Star Outer Crust Equation of State: A Machine Learning approach
• Optimization of generator coordinate method with machine-learning techniques for nuclear spectra and neutrinoless double-beta decay: ridge regression for nuclei with axial deformation
• A Kohn-Sham Scheme Based Neural Network for Nuclear Systems
• Machine learning light hypernuclei
• Building Surrogate Models of Nuclear Density Functional Theory with Gaussian Processesand Autoencoders
• Controlling extrapolations of nuclear properties with feature selection
• Machine-learning approach to finite-size effects in systems with strongly interacting fermions
• Impact of quadrupole deformation on intermediate-energy heavy-ion collisions
• A Neural Network Approach for Orienting Heavy-Ion Collision Events
• Neural Network Emulation of Flow in Heavy-Ion Collisions at Intermediate Energies
• Nuclear mass predictions based on deep neural network and finite-range droplet model (2012)
• Hybrid neural network method of a multilayer perceptron and autoencoder for the α-particle preformation factor in α-decay theory
• A Bayesian Analysis of Nuclear Deformation Properties with Skyrme Energy Functionals
• Machine learning the deuteron: new architectures and uncertainty quantification
• Report from the A.I. For Nuclear Physics Workshop
• Multimodal fission from self-consistent calculations
• Predicting the Masses of Exotic Hadrons with Data Augmentation Using Multilayer Perceptron
• Microscopic derivation of the interacting boson model parameters with machine learning
• Impact of nuclear masses on r-process nucleosynthesis: bulk properties versus shell effects