Professor Mark Tuckerman and postdoctoral fellow Leslie Vogt worked with colleagues in Germany and California at the Institute for Applied Mathematics (IPAM) to develop a way to identify patterns of molecular behavior and translate them into predictions.
"Bypassing the Kohn-Sham Equations with Machine Learning" was published in Nature Communications, and was picked up by NYU RESEARCH in a piece called, "Scientists Develop Machine-Learning Method to Predict the Behavior of Molecules."
To read the article and for a complete list of authors, click here.
To read the research highlight, click here.
Abstract: Last year, at least 30,000 scientific papers used the Kohn–Sham scheme of density functional theory to solve electronic structure problems in a wide variety of scientific fields. Machine learning holds the promise of learning the energy functional via examples, bypassing the need to solve the Kohn–Sham equations. This should yield substantial savings in computer time, allowing larger systems and/or longer time-scales to be tackled, but attempts to machine-learn this functional have been limited by the need to find its derivative. The present work overcomes this difficulty by directly learning the density-potential and energy-density maps for test systems and various molecules. We perform the first molecular dynamics simulation with a machine-learned density functional on malonaldehyde and are able to capture the intramolecular proton transfer process. Learning density models now allows the construction of accurate density functionals for realistic molecular systems.
This work was supported by the National Science Foundation, the Einstein Foundation, the Korean government and United States Army Research grants.