Simultaneous improvements of nuclear mass and charge radius predictions using multitask Gaussian process approaches
Phys. Rev. C 113, 024304 – Published 4 February, 2026
DOI: https://doi.org/10.1103/1mgv-jypl
Abstract
A multitask Gaussian process (GP) machine learning model is introduced to simultaneously predict two important nuclear observables across the nuclear chart, namely nuclear masses and charge radii. Utilizing 12 physical input features, our multitask GP consistently outperforms single-task learning. We obtain overall root-mean-square deviations of 0.136 MeV for masses and 0.007 fm for charge radii, respectively. The good performance of the present model is confirmed by three complementary validations in the present work. More explicitly, we employ the fractions of training and testing data, extrapolations to newly reported nuclei far from stability, and comparisons with the popular Garvey-Kelson mass relations. The correlations between the two observables are explicitly analyzed within the multi-task learning framework. Furthermore, by employing the SHapley Additive exPlanations (SHAP) method, we interpret the importance of different features for mass and radius predictions across distinct nuclear regions. These results demonstrate the effectiveness of the multitask GP approach for high-accuracy nuclear property predictions.
