Nuclear mass predictions with machine learning reaching the accuracy required by $r$-process studies

Abstract

Nuclear masses are predicted with the Bayesian neural networks by learning the mass surface of even-even nuclei and the correlation energies to their neighbouring nuclei. By keeping the known physics in various sophisticated mass models and performing the delicate design of neural networks, the proposed Bayesian machine learning (BML) mass model achieves an accuracy of $84$~keV, which crosses the accuracy threshold of the $100$~keV in the experimentally known region. It is also demonstrated the corresponding uncertainties of mass predictions are properly evaluated, while the uncertainties increase by about $50$~keV each step along the isotopic chains towards the unknown region. The shell structures in the known region are well described and several important features in the unknown region are predicted, such as the new magic numbers around $N=40$, the robustness of $N=82$ shell, the quenching of $N=126$ shell, and the smooth separation energies around $N=104$.

arXiv comment14 pages, 5 figures

References in corpus (40)

• The r-process of stellar nucleosynthesis: Astrophysics and nuclear physics achievements and mysteries
• A kilonova as the electromagnetic counterpart to a gravitational-wave source
• Origin of the Heaviest Elements: the Rapid Neutron-Capture Process
• Microscopic mass formulae
• Surface diffuseness correction in global mass formula
• Skyrme-Hartree-Fock-Bogoliubov nuclear mass formulas: Crossing the 0.6 MeV threshold with microscopically deduced pairing
• Identification of strontium in the merger of two neutron stars
• The impact of individual nuclear properties on $r$-process nucleosynthesis
• Collapsars as a major source of r-process elements
• 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
• Masses, Deformations and Charge Radii--Nuclear Ground-State Properties in the Relativistic Mean Field Model
• Nuclear Mass Predictions for the Crustal Composition of Neutron Stars: A Bayesian Neural Network Approach
• Current Status of r-Process Nucleosynthesis
• Bayesian approach to model-based extrapolation of nuclear observables
• New fission fragment distributions and r-process origin of the rare-earth elements
• Neutron drip line in the Ca region from Bayesian model averaging
• Dawning of the N=32 shell closure seen through precision mass measurements of neutron-rich titanium isotopes
• Bayesian Evaluation of Incomplete Fission Yields
• Impact of nuclear mass uncertainties on the $r$-process
• Taming nuclear complexity with a committee of multilayer neural networks
• First direct mass measurements of nuclides around $Z=100$ with a Multireflection Time-of-Flight Mass Spectrograph
• Radial basis function approach in nuclear mass predictions
• Precision mass measurements on neutron-rich rare-earth isotopes at JYFLTRAP - reduced neutron pairing and implications for the $r$-process calculations
• Precision Mass Measurement of ${}^{58-63}$Cr: Nuclear Collectivity towards the \emph{N}=40 Island of Inversion
• Isochronous mass measurements of $T_z=-1$ $fp$-shell nuclei from projectile fragmentation of ${}^{58}$Ni
• High precision nuclear mass predictions towards a hundred kilo-electron-volt accuracy
• First glimpse of the $N=82$ shell closure below $Z=50$ from masses of neutron-rich cadmium isotopes and isomers
• Improved radial basis function approach with the odd-even corrections
• Masses of neutron-rich ${}^{52-54}$Sc and ${}^{54,56}$Ti nuclides: The $N=32$ subshell closure in scandium
• Exploring the mass surface near the rare-earth abundance peak via precision mass measurements at JYFLTRAP
• High-precision mass measurement of ${}^{56}$Cu and the redirection of the rp-process flow
• Masses of ground and isomeric states of ${}^{101}$In and configuration-dependent shell evolution in odd-$A$ indium isotopes
• Mass Measurements of Neutron-Rich Gallium Isotopes Refine Production of Nuclei of the First r-Process Abundance Peak in Neutron Star Merger Calculations
• Examining the $N$ = 28 shell closure through high-precision mass measurements of ${}^{46-48}$Ar
• High-precision mass measurements for the isobaric multiplet mass equation at A = 52
• Precision Mass Measurements of Neutron-Rich Co Isotopes Beyond N=40
• Study of the long-lived excited state in the neutron deficient nuclides ${}^{195,197,199}$Po by precision mass measurement
• Precision mass measurements of magnesium isotopes and implications on the validity of the Isobaric Mass Multiplet Equation
• High precision mass measurements of the isomeric and ground states of ${}^{44}$V: improving constraints on the IMME parameters of the A=44, ${\text{0}}^{\text{+}}$ quintet

Cited by in corpus (17)

• Prediction of Nuclear Charge Density Distribution with Feedback Neural Network
• Principal components of nuclear mass models
• Skyrme-Hartree-Fock-Bogoliubov mass models on a 3D mesh. IIb. Fission properties of BSkG2
• Nuclear mass predictions based on convolutional neural network
• 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
• Binding energies of ground and isomeric states in neutron-rich ruthenium isotopes: measurements at JYFLTRAP and comparison to theory
• Correlation between U/Th and Pb/Os abundance ratios and its application in nuclear cosmochronology
• Re examination of the effect of Pairing Gaps on Gamow Teller Strength Distributions and beta decay Rates
• Inference of Parameters for Back-shifted Fermi Gas Model using Feedback Neural Network
• Nuclear mass predictions based on deep neural network and finite-range droplet model (2012)
• Shell quenching in nuclear charge radii based on Monte Carlo dropout Bayesian neural network
• Investigating the effects of precise mass measurements of Ru and Pd isotopes on machine learning mass modeling
• Bayesian model mixing with multi-reference energy density functional
• Exploratory study on the masses of odd-$Z$ nuclei and $r$-process simulation based on the deformed relativistic Hartree-Bogoliubov theory in continuum
• Proton radioactivity in deformed nuclei with microscopic optical potential: A novel angular-dependent emission mechanism in the nanosecond-lived ${}^{149}$Lu
• Principal Components of Nuclear Mass Model Residuals