Abstract:Background: The generator coordinate method (GCM) is an important tool of choice for modeling large-amplitude collective motion in atomic nuclei. Recently, it has attracted increasing interest as it can be exploited to extend ab initio methods to the collective excitations of medium-mass and heavy deformed nuclei, as well as the nuclear matrix elements (NME) of candidates for neutrinoless double-beta ($0\ensuremath{\nu}\ensuremath{\beta}\ensuremath{\beta}$) decay.Purpose: The computational complexity of the GCM increases rapidly with the number of collective coordinates. It imposes a strong restriction on the applicability of the method. We aim to exploit statistical machine-learning (ML) algorithms to speed up GCM calculations and ultimately provide a more efficient description of nuclear energy spectra and other observables such as the NME of $0\ensuremath{\nu}\ensuremath{\beta}\ensuremath{\beta}$ decay without loss of accuracy.Method: In this work, we propose a subspace-reduction algorithm that employs optimal statistical ML models as surrogates for exact quantum-number projection calculations for norm and Hamiltonian kernels. The model space of the original GCM is reduced to a subspace relevant for nuclear low energy spectra and the NME of ground state to ground state $0\ensuremath{\nu}\ensuremath{\beta}\ensuremath{\beta}$ decay based on the orthogonality condition (OC) and the energy-transition-orthogonality procedure (ENTROP), respectively. Nuclear energy spectra are determined by the GCM through the configuration mixing within this subspace. For simplicity, the polynomial ridge regression (RR) algorithm is used to learn the norm and Hamiltonian kernels of axially deformed configurations. The efficiency and accuracy of this algorithm are illustrated for $^{76}\mathrm{Ge}$ and $^{76}\mathrm{Se}$ by comparing results obtained using the optimal RR models to direct GCM calculations. The nonrelativistic Gogny force D1S and relativistic energy density functional PC-PK1, a valence-space shell-model Hamiltonian, and a modern nuclear interaction derived from chiral effective field theory are employed.Results: The low-lying energy spectra of $^{76}\mathrm{Ge}$ and $^{76}\mathrm{Se}$, as well as the $0\ensuremath{\nu}\ensuremath{\beta}\ensuremath{\beta}$-decay NME between their ground states, are computed. The results show that the performance of the $\mathrm{GCM}+\text{OC/ENTROP}+\mathrm{RR}$ is more robust than that of the $\mathrm{GCM}+\mathrm{RR}$ alone, and the former can reproduce the results of the original GCM calculation accurately with a significantly reduced computational cost.Conclusions: Statistical ML algorithms, when implemented properly, can accelerate GCM calculations without loss of accuracy. In applications with axially deformed states, the computation time can be reduced by a factor of three to nine for energy spectra and NMEs, respectively. This factor is expected to increase significantly with the number of employed generator coordinates.

Neutrinoless double-$\beta$ decay of $^{124}$Sn, $^{130}$Te, and $^{136}$Xe in the Hamiltonian-based generator-coordinate method

Neutrinoless Double- Β Decay of Sn124 , Te130 , And

Neutrinoless Double-Β Decay Matrix Elements in Large Shell-Model Spaces with the Generator-Coordinate Method

Generator-coordinate Reference States for Spectra and 0νββ Decay in the In-Medium Similarity Renormalization Group

Union of Rotational and Vibrational Modes in Generator-Coordinate-type Calculations, with Application to Neutrinoless Double- Β Decay

Novel shell-model analysis of the $^{136}$Xe double beta decay nuclear matrix elements

Nuclear matrix elements calculation for $0νββ$ decay of $^{124}$Sn using nonclosure approach in nuclear shell model

Nuclear shell model study of neutrinoless double beta decay under Left-Right symmetric model

Comparison between Variational Monte Carlo and Shell Model Calculations of Neutrinoless Double Beta Decay Matrix Elements in Light Nuclei

Nuclear matrix elements of neutrinoless double-beta decay in covariant density functional theory with different mechanisms

Shell model analysis of competing mechanism to the double-beta decay of $^{48}$Ca

Optimization of the generator coordinate method with machine-learning techniques for nuclear spectra and neutrinoless double- <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mi>β</mml:mi></mml:math> decay: Ridge regression for nuclei with axial deformation

Large-scale shell-model study of two-neutrino double-beta decay of $^{82}$Se, $^{94}$Zr, $^{108}$Cd, $^{124}$Sn, $^{128}$Te, $^{130}$Te, $^{136}$Xe, and $^{150}$Nd

Impact of the leading-order short-range nuclear matrix element on the neutrinoless double-beta decay of medium-mass and heavy nuclei

Deformed shell model results for neutrinoless double beta decay of nuclei in A=60-90 region

Shell-model calculation of $^{100}$Mo double-$β$ decay

Optimization of generator coordinate method with machine-learning techniques for nuclear spectra and neutrinoless double-beta decay: ridge regression for nuclei with axial deformation

Nuclear Matrix Elements of Neutrinoless Double-Β Decay in the Triaxial Projected Shell Model

Study of Nuclear Effects in the Computation of the 0νββ Decay Matrix Elements

Neutrinoless double-beta decay in the neutrino-extended Standard Model

Probing new physics in the Neutrinoless double beta decay using electron angular correlation