Background: Accurate knowledge of fission fragment yields is an essential ingredient of numerous applications ranging from the formation of elements in the process to fuel cycle optimization for nuclear energy. The need for a predictive theory applicable where no data are available, together with the variety of potential applications, is an incentive to develop a fully microscopic approach to fission dynamics.
26 April 2022 in Progress in Particle and Nuclear Physics
We demonstrate that a committee of deep neural networks is capable of predicting the ground-state and excited energies of more than 1800 atomic nuclei with an accuracy akin to the one achieved by state-of-the-art nuclear energy density functionals (EDFs) and with significantly less computational cost. An active learning strategy is proposed to train this algorithm with a minimal set of 210 nuclei. This approach enables future fast studies of the influence of EDF parametrizations on structure properties over the whole nuclear chart and suggests that for the first time a machine learning framework successfully encoded several correlated aspects of nuclear deformation.
Background: Predicting the properties of neutron-rich nuclei far from the valley of stability is one of the major challenges of modern nuclear theory. In heavy and superheavy nuclei, a difference of only a few neutrons is sufficient to change the dominant fission mode. A theoretical approach capable of predicting such rapid transitions for neutron-rich systems would be a valuable tool to better understand -process nucleosynthesis or the decay of superheavy elements.