Structure-optimized potential refinement (SOPR) is a machine learning assisted iterative Boltzmann inversion method designed to predict accurate and transferable interaction potentials from a provided set of site-site partial radial distribution functions.

Structure and self-assembly are complex, emergent properties of matter that are often misrepresented by existing molecular simulation models. In this project, we use Bayesian optimization, an accurate and robust statistical method, to optimize novel force fields based on experimental neutron/X-ray diffraction data to better model the structural behavior of liquid state systems.

Gaussian processes provide a Bayesian framework to analyze experimental scattering data while performing rigorous uncertainty quantification and propagation (UQ/P). Our team uses state-of-the-art Gaussian process approaches, including spectral and non-stationary kernel design, to design next generation scattering analysis tools.