Figures involved in the manuscript entitled "Physics-Embedded Bayesian Neural Network (PE-BNN) to predict Energy Dependence of Fission Product Yields with Fine Structures"

We present a physics-embedded Bayesian neural network (PE-BNN) framework that integratesfission product yields (FPYs) with prior nuclear physics knowledge to predict energy-dependentFPY data with fine structure. By incorporating a phenomenological shell-related input feature withan excitation-energy damping term, the PE-BNN captures both fine structures and global energytrends. The combination of this physics-informed input with hyperparameter optimization via theWatanabe-Akaike Information Criterion (WAIC) significantly enhances predictive performance. Ourresults show that the PE-BNN framework is well suited to observables whose systematic featurescan be encoded as model inputs, and that it remains consistent with known shell-related trends andindependently measured prompt-neutron systematics.