Title:NucleiML: A machine learning framework of ground-state properties of finite nuclei for accelerated Bayesian exploration

Abstract:The global behavior of the nuclear equation of state (EoS) is usually investigated using finite nuclei (FN) data, along with constraints from heavy-ion collisions and astrophysical observations of neutron star (NS) properties. The FN constraints explicitly imposed through the binding energies and charge radii of selected nuclei are found to significantly affect the EoS across different densities. However, the high computational cost of these constraints makes it challenging to extend the analysis to a broader set of nuclei, particularly when the objective is not merely to obtain a single optimized model but to systematically explore uncertainties in global modeling. To overcome this challenge, we introduce NucleiML (NML), a machine learning framework trained on ground-state FN properties obtained from mean-field models. Integrated into a Bayesian inference approach, NML demonstrates high accuracy and strong consistency with the underlying mean-field model. The NML achieves around ten-fold computational speed-up, from $\sim 4.5$ hours to 30 minutes. Its predictive performance improves further as the number of nuclei in the training data increases, which we plan to employ in extensive future explorations.