Mixture density network in evaluating incomplete fission mass yields

Abstract

Accurately modeling fission product yields (FPY) is crucial yet challenging due to the complex quantum-mechanical nature of nuclear reactions. Traditional models face limitations in predictive power and handling evolving fission modes. Neural Networks (NNs) present a promising solution to these challenges by effectively modeling and predicting energy-dependent fission yields. Mixture Density Networks (MDNs) enable learning from available data, predicting unknowns, and quantifying uncertainties simultaneously. Machine learning algorithms like Gaussian Process Regression (GPR) can capture the distribution of single-fission yields and generate high-quality samples. These samples serve as valuable inputs for MDN networks. This study introduces an MDN approach for evaluating energy-dependent fission mass yields. The results indicate satisfactory accuracy in determining both the distribution positions and energy dependencies of FPYs, particularly in scenarios where experimental data are incomplete.