Adaptive Machine Learning enabled Search for Functional Materials with Targeted Properties

P.V. Balachandran
University of Virginia,
United States

Keywords: materials, machine learning, uncertainties, active learning, domain knowledge, density functional theory


Traditionally, new materials are discovered by intuition and trial-and-error approaches. However, as the material complexities increase, the combinatorial possibilities become too large for such approaches to be practical. Computational strategies that enable efficient navigation of this vast search space have the potential to accelerate the search and discovery of novel materials. In this talk, I will discuss some of our recent works that have rationally guided experiments and computational codes (eg., density functional theory) towards promising regions in the vast design space of functional materials.