Teaching Computers Chemistry
Machine Learning (ML) methods seek to use a variety of statistical tools, including neural networks, to analyze and train on massive data sets. Through both supervised and active learning methods, these ML models become increasingly accurate predictors of both computational and experimental data.
We draw on ML methods in our materials discovery research, seeking to meet the goal of accurate quantum chemical predictions with much faster speed than conventional methods that require hours or days of calculations.
Recent Publications:
Screening Efficient Tandem Organic Solar Cells with Machine Learning and Genetic Algorithms
Brianna L. Greenstein, Geoffrey R. Hutchison. “Screening Efficient Tandem Organic Solar Cells with Machine Learning and Genetic Algorithms” J. Phys. Chem. C ...
Deep learning coordinate-free quantum chemistry
Matthew K Matlock, Max Hoffman, Na Le Dang, Dakota L. Folmsbee, Luke A. Langkamp, Geoffrey R. Hutchison, Neeraj Kumar, and S. Joshua Swamidass.”Deep learning...
Machine learning to accelerate screening for Marcus reorganization energies
Omri D. Abarbanel, Geoffrey R. Hutchison. “Machine learning to accelerate screening for Marcus reorganization energies” J. Chem. Phys. 155, 054106 (2021). D...
Evaluation of Thermochemical Machine Learning for Potential Energy Curves and Geometry Optimization
Dakota L. Folmsbee, David R. Koes, Geoffrey R. Hutchison. “Evaluation of Thermochemical Machine Learning for Potential Energy Curves and Geometry Optimizatio...