Bayesian approach to model-based extrapolation of nuclear observables

Abstract

The mass, or binding energy, is the basis property of the atomic nucleus. It determines its stability, and reaction and decay rates. Quantifying the nuclear binding is important for understanding the origin of elements in the universe. The astrophysical processes responsible for the nucleosynthesis in stars often take place far from the valley of stability, where experimental masses are not known. In such cases, missing nuclear information must be provided by theoretical predictions using extreme extrapolations. Bayesian machine learning techniques can be applied to improve predictions by taking full advantage of the information contained in the deviations between experimental and calculated masses. We consider 10 global models based on nuclear Density Functional Theory as well as two more phenomenological mass models. The emulators of S2n residuals and credibility intervals defining theoretical error bars are constructed using Bayesian Gaussian processes and Bayesian neural networks. We consider a large training dataset pertaining to nuclei whose masses were measured before 2003. For the testing datasets, we considered those exotic nuclei whose masses have been determined after 2003. We then carried out extrapolations towards the 2n dripline. While both Gaussian processes and Bayesian neural networks reduce the rms deviation from experiment significantly, GP offers a better and much more stable performance. The increase in the predictive power is quite astonishing: the resulting rms deviations from experiment on the testing dataset are similar to those of more phenomenological models. The empirical coverage probability curves we obtain match very well the reference values which is highly desirable to ensure honesty of uncertainty quantification, and the estimated credibility intervals on predictions make it possible to evaluate predictive power of individual models.

References in corpus (38)

• Nuclear ground-state masses and deformations: FRDM(2012)
• Microscopic mass formulae
• Relativistic Nuclear Energy Density Functionals: adjusting parameters to binding energies
• Nuclear Energy Density Optimization
• Structure of even-even nuclei using a mapped collective Hamiltonian and the D1S Gogny interaction
• Nuclear energy density optimization: Large deformations
• The effective force NL3 revisited
• Error Estimates of Theoretical Models: a Guide
• Global performance of covariant energy density functionals: ground state observables of even-even nuclei and the estimate of theoretical uncertainties
• Nuclear mass predictions based on Bayesian neural network approach with pairing and shell effects
• The limits of the nuclear landscape explored by the relativistic continuum Hatree-Bogoliubov theory
• Nuclear Mass Predictions for the Crustal Composition of Neutron Stars: A Bayesian Neural Network Approach
• Nuclear energy density optimization: Shell structure
• Relativistic mean field interaction with density dependent meson-nucleon vertices based on microscopical calculations
• Global study of quadrupole correlation effects
• One-quasiparticle States in the Nuclear Energy Density Functional Theory
• Nuclear charge radii: Density functional theory meets Bayesian neural networks
• Uncertainty Quantification for Nuclear Density Functional Theory and Information Content of New Measurements
• The neutron drip line: single-particle degrees of freedom and pairing properties as sources of theoretical uncertainties
• Refining mass formulas for astrophysical applications: a Bayesian neural network approach
• Nuclear mass systematics using neural networks
• Covariant energy density functionals: nuclear matter constraints and global ground state properties
• Systematics on ground-state energies of nuclei within the neural networks
• Propagation of uncertainties in the Skyrme energy-density-functional model
• Precision mass measurements on neutron-rich rare-earth isotopes at JYFLTRAP - reduced neutron pairing and implications for the $r$-process calculations
• Global microscopic calculations of ground-state spin and parity for odd-mass nuclei
• Validating neural-network refinements of nuclear mass models
• Gamow-Hartree-Fock-Bogoliubov Method: Representation of quasiparticles with Berggren sets of wave functions
• Positioning the neutron drip line and the r-process paths in the nuclear landscape
• Uncertainty decomposition method and its application in the liquid drop model
• Estimating parameter uncertainty in binding-energy models by the frequency-domain bootstrap
• Uncertainty Quantification and Propagation in Nuclear Density Functional Theory
• Fluctuating parts of nuclear ground state correlation energies
• Optimizing relativistic energy density functionals: covariance analysis
• Propagation of uncertainties in the nuclear DFT models
• Addressing spectroscopic quality of covariant density functional theory
• Uncertainty propagation within the UNEDF models
• Hartree-Fock-Bogoliubov solution of the pairing Hamiltonian in finite nuclei

Cited by in corpus (84)

• Origin of the Heaviest Elements: the Rapid Neutron-Capture Process
• Machine Learning in Nuclear Physics
• Ab initio limits of atomic nuclei
• Future of Nuclear Fission Theory
• Neutron drip line in the Ca region from Bayesian model averaging
• Predictions of nuclear $\beta$-decay half-lives with machine learning and their impacts on $r$ process
• Machine learning the nuclear mass
• Bayesian Evaluation of Incomplete Fission Yields
• Nuclear mass predictions with machine learning reaching the accuracy required by $r$-process studies
• Regressive and generative neural networks for scalar field theory
• Quantified limits of the nuclear landscape
• Get on the BAND Wagon: A Bayesian Framework for Quantifying Model Uncertainties in Nuclear Dynamics
• Taming nuclear complexity with a committee of multilayer neural networks
• Study of Charge Radii with Neural Networks
• Direct comparison between Bayesian and frequentist uncertainty quantification for nuclear reactions
• Towards grounding nuclear physics in QCD
• Opportunities for Fundamental Physics Research with Radioactive Molecules
• Deep learning: Extrapolation tool for ab initio nuclear theory
• Impact of the neutron star crust on the tidal polarizability
• Nuclear masses learned from a probabilistic neural network
• Nuclear masses in extended kernel ridge regression with odd-even effects
• Multi-task learning on nuclear masses and separation energies with the kernel ridge regression
• Physically Interpretable Machine Learning for nuclear masses
• Beyond the proton drip line: Bayesian analysis of proton-emitting nuclei
• Extrapolation of nuclear structure observables with artificial neural networks
• Skyrme-Hartree-Fock-Bogoliubov mass models on a 3D Mesh: Effect of triaxial shape
• Nuclear energy density functionals from machine learning
• Two-proton emission and related phenomena
• Toward emulating nuclear reactions using eigenvector continuation
• Bayesian evaluation of charge yields of fission fragments of 239U
• Propagation of Statistical Uncertainties of Skyrme Mass Models to Simulations of $r$-Process Nucleosynthesis
• Statistical aspects of nuclear mass models
• Nuclear mass predictions using machine learning models
• Towards Precise and Accurate Calculations of Neutrinoless Double-Beta Decay: Project Scoping Workshop Report
• Nuclear binding energy predictions using neural networks: Application of the multilayer perceptron
• Prediction of Nuclear Charge Density Distribution with Feedback Neural Network
• Lightweight self-conjugate nucleus ${}^{80}$Zr
• Quantifying Uncertainties on Fission Fragment Mass Yields With Mixture Density Networks
• Improved phenomenological nuclear charge radius formulae with kernel ridge regression
• Sensitivity study of \emph{r}-process abundances to nuclear masses
• The description of giant dipole resonance key parameters with multitask neural networks
• Recent advances in the quantification of uncertainties in reaction theory
• Alpha Decay Energies of Superheavy Nuclei: Systematic Trends
• Electroweak probes of ground state densities
• Nuclear charge radius predictions by kernel ridge regression with odd-even effects
• Emergence of a pseudogap in the BCS-BEC crossover
• Impact of statistical uncertainties on the composition of the outer crust of a neutron star
• Optimizing multilayer Bayesian neural networks for evaluation of fission yields
• Exploring experimental conditions to reduce uncertainties in the optical potential
• Providing physics guides in Bayesian neural networks from input layer: case of giant dipole resonance predictions
• A Bayesian-Neural-Network Prediction for Fragment Production in Proton Induced Spallation Reaction
• Trees and Forests in Nuclear Physics
• Machine learning approach to pattern recognition in nuclear dynamics from the ab initio symmetry-adapted no-core shell model
• Markov Chain Monte Carlo Predictions of Neutron-rich Lanthanide Properties as a Probe of $r$-process Dynamics
• Extrapolating from neural network models: a cautionary tale
• Principal components of nuclear mass models
• Nuclear mass predictions based on convolutional neural network
• Skyrme-Hartree-Fock-Bogoliubov mass models on a 3D mesh. IIb. Fission properties of BSkG2
• Application of multilayer perceptron with data augmentation in nuclear physics
• Non-parametric Bayesian approach to extrapolation problems in configuration interaction methods
• Optimization of generator coordinate method with machine-learning techniques for nuclear spectra and neutrinoless double-beta decay: ridge regression for nuclei with axial deformation
• Kolmogorov-Arnold networks in nuclear binding energy prediction
• Building Surrogate Models of Nuclear Density Functional Theory with Gaussian Processesand Autoencoders
• Machine learning light hypernuclei
• Controlling extrapolations of nuclear properties with feature selection
• Calibration of Energy Density Functionals with Deformed Nuclei
• Classical-Quantum Noise Mitigation for NISQ Hardware
• Interpolating between small- and large-$g$ expansions using Bayesian Model Mixing
• Does Bayesian Model Averaging improve polynomial extrapolations? Two toy problems as tests
• A Fast and Calibrated Computer Model Emulator: An Empirical Bayes Approach
• Towards accurate nuclear mass tables in covariant density functional theory
• Discovering Nuclear Models from Symbolic Machine Learning
• The nucleardatapy toolkit for simple access to experimental nuclear data, astrophysical observations, and theoretical predictions
• Shell quenching in nuclear charge radii based on Monte Carlo dropout Bayesian neural network
• A Bayesian Analysis of Nuclear Deformation Properties with Skyrme Energy Functionals
• Quantifying uncertainty in machine learning on nuclear binding energy
• Ab initio models of atomic nuclei: challenges and new ideas
• Report from the A.I. For Nuclear Physics Workshop
• Exploratory study on the masses of odd-$Z$ nuclei and $r$-process simulation based on the deformed relativistic Hartree-Bogoliubov theory in continuum
• Complex-scaled no-core shell model calculations of bound and unbound nuclear states in light nuclei
• FRIB and the GW170817 Kilonova
• Gaussian process error modeling for chiral effective-field-theory calculations of $np\leftrightarrow d\gamma$ at low energies
• Statistical correlations of nuclear quadrupole deformations and charge radii
• Principal Components of Nuclear Mass Model Residuals