Uncertainties here, there, and everywhere: Bayesian uncertainty quantification and model mixing in nuclear physics

As nuclear physics reaches higher levels of precision experimentally and theoretically, the quantification of model uncertainties has become critical for improved theoretical predictions. Bayesian methods allow us to perform this uncertainty quantification (UQ) of individual models, as well as make new predictions from combinations of models using Bayesian model mixing (BMM). BMM is essential to capitalize on the information from several models describing the same system, or to connect between models that do not overlap. In this talk I will review the advances made in the past several years to perform UQ on individual models in nuclear physics, specifically quantifying the truncation error of effective field theories (EFTs) using Gaussian processes. I will also discuss several general techniques to perform BMM, with a focus on applying these methods to the dense matter equation of state for neutron stars. I will conclude with a brief discussion of potential future applications of model UQ and BMM in the fields of nuclear structure and reaction theory.