Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning

TLDR

We introduce a machine learning model to predict atomization energies of diverse organic molecules using only nuclear charges and atomic positions. The model maps the Schrödinger equation to a nonlinear regression problem, training on hybrid DFT‑computed atomization energies of diverse organic molecules using only nuclear charges and atomic positions. Cross‑validation on over 7,000 molecules shows a mean absolute error of ~10 kcal/mol, and the model can predict atomization potential energy curves.

Abstract

We introduce a machine learning model to predict atomization energies of a diverse set of organic molecules, based on nuclear charges and atomic positions only. The problem of solving the molecular Schrödinger equation is mapped onto a nonlinear statistical regression problem of reduced complexity. Regression models are trained on and compared to atomization energies computed with hybrid density-functional theory. Cross validation over more than seven thousand organic molecules yields a mean absolute error of ∼10 kcal/mol. Applicability is demonstrated for the prediction of molecular atomization potential energy curves.