Nuclear masses learned from a probabilistic neural network

Abstract

Machine learning methods and uncertainty quantification have been gaining interest throughout the last several years in low-energy nuclear physics. In particular, Gaussian processes and Bayesian Neural Networks have increasingly been applied to improve mass model predictions while providing well-quantified uncertainties. In this work, we use the probabilistic Mixture Density Network (MDN) to directly predict the mass excess of the 2016 Atomic Mass Evaluation within the range of measured data, and we extrapolate the inferred models beyond available experimental data. The MDN not only provides mean values but also full posterior distributions both within the training set and extrapolated testing set. We show that the addition of physical information to the feature space increases the accuracy of the match to the training data as well as provides for more physically meaningful extrapolations beyond the the limits of experimental data.

arXiv comment10 pages, 3 figures, under review with Phys. Rev. C

References in corpus (41)

• Nuclear ground-state masses and deformations: FRDM(2012)
• Origin of the Heaviest Elements: the Rapid Neutron-Capture Process
• Microscopic mass formulae
• Skyrme-Hartree-Fock-Bogoliubov nuclear mass formulas: Crossing the 0.6 MeV threshold with microscopically deduced pairing
• The impact of individual nuclear properties on $r$-process nucleosynthesis
• Microscopic Theory of Nuclear Fission: A Review
• Quantifying truncation errors in effective field theory
• Nuclear Mass Predictions for the Crustal Composition of Neutron Stars: A Bayesian Neural Network Approach
• Ab initio limits of atomic nuclei
• Further improvements on a global nuclear mass model
• Bayesian approach to model-based extrapolation of nuclear observables
• r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
• Bayesian truncation errors in chiral effective field theory: nucleon-nucleon observables
• Neutron drip line in the Ca region from Bayesian model averaging
• Determining Fundamental Properties of Matter Created in Ultrarelativistic Heavy-Ion Collisions
• Nuclear charge radii: Density functional theory meets Bayesian neural networks
• Eigenvector continuation with subspace learning
• Uncertainty Quantification for Nuclear Density Functional Theory and Information Content of New Measurements
• Global sensitivity analysis of bulk properties of an atomic nucleus
• Bayesian parameter estimation for effective field theories
• Bayesian Evaluation of Incomplete Fission Yields
• Eigenvector Continuation as an Efficient and Accurate Emulator for Uncertainty Quantification
• 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
• Rigorous constraints on three-nucleon forces in chiral effective field theory from fast and accurate calculations of few-body observables
• Direct comparison between Bayesian and frequentist uncertainty quantification for nuclear reactions
• Statistical Hauser-Feshbach theory with width fluctuation correction including direct reaction channels for neutron induced reaction at low energies
• Bayesian Methods for Parameter Estimation in Effective Field Theories
• Efficient emulators for scattering using eigenvector continuation
• Towards a Deeper Understanding of How Experiments Constrain the Underlying Physics of Heavy-Ion Collisions
• Beyond the proton drip line: Bayesian analysis of proton-emitting nuclei
• Effective field theory for nuclear vibrations with quantified uncertainties
• Convergence of Eigenvector Continuation
• Constraining Transfer Cross Sections Using Bayes' Theorem
• Quantifying Uncertainties on Fission Fragment Mass Yields With Mixture Density Networks
• Uncertainty Quantification for Optical Model Parameters
• Optimizing multilayer Bayesian neural networks for evaluation of fission yields
• Uncertainty Quantification and Propagation in Nuclear Density Functional Theory
• Uncertainty quantification due to optical potentials in models for (d,p) reactions
• Markov Chain Monte Carlo Predictions of Neutron-rich Lanthanide Properties as a Probe of $r$-process Dynamics

Cited by in corpus (17)

• Machine Learning in Nuclear Physics
• Physically Interpretable Machine Learning for nuclear masses
• Nuclear mass predictions using machine learning models
• Nuclear mass predictions based on convolutional neural network
• Modeling Heavy-Ion Fusion Cross Section Data via a Novel Artificial Intelligence Approach
• Kolmogorov-Arnold networks in nuclear binding energy prediction
• Building Surrogate Models of Nuclear Density Functional Theory with Gaussian Processesand Autoencoders
• High Precision Binding Energies from Physics Informed Machine Learning
• Neural Network Emulation of Flow in Heavy-Ion Collisions at Intermediate Energies
• Shell quenching in nuclear charge radii based on Monte Carlo dropout Bayesian neural network
• Quantifying uncertainty in machine learning on nuclear binding energy
• Investigating the effects of precise mass measurements of Ru and Pd isotopes on machine learning mass modeling
• Iso-scalar liquid drop model
• Bayesian model mixing with multi-reference energy density functional
• Further exploration of binding energy residuals using machine learning and the development of a composite ensemble model
• Nuclear data activities for medium mass and heavy nuclei at Los Alamos
• Bayesian method for fitting the low-energy constants in chiral perturbation theory