Simultaneous improved predictions of nuclear mass and charge radius via multi-task Gaussian process machine learning approach

We introduce a multi-task Gaussian process (GP) machine learning model to simultaneously predict nuclear masses and charge radii across the nuclear chart. Utilizing 12 physical input features, our multi-task GP consistently outperforms single-task learning, achieving root-mean-square deviations of 0.09 MeV for masses and 0.004 fm for charge radii. The model explicitly incorporates correlations between these observables. By employing the SHapley Additive exPlanations (SHAP) method, we interpret the significance of different features for mass and radius predictions across distinct nuclear regions. These results demonstrate the effectiveness of the multi-task GP approach for high-accuracy nuclear property predictions.