Improvement of the Bayesian neural network to study the photoneutron yield cross sections

his work is an attempt to improve the Bayesian neural network (BNN) for studying photoneutron yield crosssections as a function of the charge number Z, mass number A, and incident energy ". The BNN was improvedin terms of three aspects: numerical parameters, input layer, and network structure. First, by minimizing thedeviations between the predictions and data, the numerical parameters, including the hidden layer number,hidden node number, and activation function, were selected. It was found that the BNN with three hiddenlayers, 10 hidden nodes, and sigmoid activation function provided the smallest deviations. Second, based onknown knowledge, such as the isospin dependence and shape effect, the optimal ground-state properties wereselected as input neurons. Third, the Lorentzian function was applied to map the hidden nodes to the outputcross sections, and the empirical formula of the Lorentzian parameters was applied to link some of the inputnodes to the output cross sections. It was found that the last two aspects improved the predictions and avoidedoverfitting, especially for the axially deformed nucleus.