Projects

Bayesic Force Fields (BFF) is an open source, Bayesian force field learning tool aimed at addressing transferability and robustness issues with existing biomolecular force fields. The code performs efficient Bayesian inference with physically motivated priors on Coulombic, bonded, and non-bonded terms.

Gaussian processes can model complex experimental data and interatomic interactions while enabling rigorous uncertainty quantification and propagation within a Bayesian framework. We apply state-of-the-art GP methods, including local [1], spectral, and non-stationary kernel design [2], to develop next-generation molecular simulation tools.

Gaussian processes (GPs) are a Bayesian regression technique with big upside for machine learning interatomic potentials. Unlike black-box methods like neural networks, GPs are interpretable, uncertainty-aware, and allow for the integration of physics-informed behavior through the use of carefully designed Bayesian priors.