Beamlines · MLIP physics · materials translation
Experiment-grounded atomistic ML for mechanistic insights and materials discovery.
Our work advances active learning, lifelong learning, and GPU-accelerated atomistic ML for beamline-validated models and translational biomedical materials discovery with dental medicine collaborators.
Our rubric
Algorithmic cores for real-world experimental practice.
Method development is treated as a feedback process, not a standalone benchmark. Beamline collaborations on technologically relevant disordered oxides and layered chalcogenides expose where models must match measured scattering signatures and atomistic structure. Translational biomedical materials projects then test whether validated models can support decisions that matter in laboratory settings.
Build scalable MLIP workflows and corrections for structures where locality alone is insufficient.
Compare simulations with beamline measurements from disordered oxides and layered chalcogenides.
Work with biomedical materials collaborators on discovery problems where atomistic mechanisms matter.
At a Glance
Research
Algorithm development for active/lifelong learning and long-range MLIPs, grounded through beamline-informed oxide and chalcogenide modeling and translated into biomedical materials discovery.
View research areas →Publications
Preprints and selected published works from 2020 onward, with the full record linked through Google Scholar.
Browse outputs →Open Software
Community-facing tools for active learning, molecular machine learning, and long-range interactions.
Explore software →