Bayesian Statistical Machine Learning for Nuclear Physics

Bayes's Theorem is a powerful tool for addressing "inverse problems" by enabling the systematic incorporation of prior knowledge and the derivation of a posterior distribution with correlations and covariances between elements of measurements and calculations. We develop novel and general Machine Learning-based approaches to Bayesian probabilistic analysis methods for application to a broad range of Nuclear Physics (NP) research areas: the mass and fundamental nature of the neutrino, study of the Quark-Gluon Plasma, and mapping of natural and anthropogenic radiation environments. I will highlight the AI/ML-powered algorithms we developed and apply them to challenging Bayesian inference calculations and computationally intensive simulations. These tools are critical for achieving the next-level Bayesian inference in Nuclear Physics with theoretical uncertainty quantification.