The Data-Driven Materials Design group develops machine-learning methods to understand, simulate, and design materials across length and time scales, from atomistic processes to larger-scale materials behavior.<\/p>\n\n\n\n

We develop machine-learning approaches for modeling materials and molecular systems, with machine-learned interatomic potentials and atomistic foundation models as central areas of our work. More broadly, our research spans the analysis and development of models, covering their accuracy, efficiency, and transferability, alongside new learning strategies and physically informed approaches. We also extend these ideas beyond atomistic simulation to materials-property prediction and data-driven materials discovery.

Working at the interface of materials science, chemistry, physics, and artificial intelligence, we develop predictive computational tools that deepen our understanding of materials and support their discovery and design.<\/p>\n<\/div>\n\n\n\n

<\/div>\n<\/div>\n\n\n\n

<\/p>\n\n\n\n

<\/div>\n\n\n