Bayesian Machine Learning for Complex Systems (BayMaLES): From point estimates to uncertainty predictions in nuclear astrophysics experiments

However, many questions remain when it comes to the creation of elements heavier than the iron group. In particular, the intermediate and the rapid neutron-capture processes remain a huge challenge to understand. Both these astrophysical processes present enormous nuclear-physics challenges as they involve very short-lived, unstable nuclei that are extremely difficult to study experimentally. Current machine learning algorithms have proven to be very effective in solving a wide variety of complex tasks. While much attention has been devoted to studying optimal solutions, many problems require a detailed knowledge of the uncertainties of predictions. In this project we work on the development of Bayesian Neural Networks and Gaussian Processes applied to obtain robust predictions from experimental data in nuclear astrophysics.