Using machine learning, a map from the external potential of a system to its electron density was trained, which allows explicit solution of the electronic Schroedinger equation to be bypassed. Using this map, a second machine learning model can be trained to obtain the total energy at quantum chemical accuracy with no requirement of self-consistency. Leslie Vogt-Maranto (Tuckerman Group postdoctoral fellow) led this study with Mihail Bogojeski (Technical University of Berlin). Other authors include Klaus-Robert Muller (TUB), Kieron Burke (UC/Irvine) and NYU Professor of Chemistry and Mathematics Mark Tuckerman.
Read the article "Quantum Chemical Accuracy from Density Functional Approximations via Machine Learning, click here.
Abstract: Kohn-Sham density functional theory (DFT) is a standard tool in most branches of chemistry, but accuracies for many molecules are limited to 2-3 kcal ⋅ mol−1 with presently-available functionals. Ab initio methods, such as coupled-cluster, routinely produce much higher accuracy, but computational costs limit their application to small molecules. In this paper, we leverage machine learning to calculate coupled-cluster energies from DFT densities, reaching quantum chemical accuracy (errors below 1 kcal ⋅ mol−1) on test data. Moreover, densitybased Δ-learning (learning only the correction to a standard DFT calculation, termed Δ-DFT ) signiﬁcantly reduces the amount of training data required, particularly when molecular symmetries are included. The robustness of Δ-DFT is highlighted by correcting “on the ﬂy” DFT-based molecular dynamics (MD) simulations of resorcinol (C6H4(OH)2) to obtain MD trajectories with coupled-cluster accuracy. We conclude, therefore, that Δ-DFT facilitates running gas-phase MD simulations with quantum chemical accuracy, even for strained geometries and conformer changes where standard DFT fails.
The NYU research was supported by the U.S. Army Research Office.