OrbNet: Deep learning for quantum chemistry using symmetry-adapted atomic-orbital features
Abstract
We introduce a machine learning method in which energy solutions from the Schrödinger equation are predicted using symmetry adapted atomic orbital features and a graph neural-network architecture. OrbNet is shown to outperform existing methods in terms of learning efficiency and transferability for the prediction of density functional theory results while employing low-cost features that are obtained from semi-empirical electronic structure calculations. For applications to datasets of drug-like molecules, including QM7b-T, QM9, GDB-13-T, DrugBank, and the conformer benchmark dataset of Folmsbee and Hutchison [Int. J. Quantum Chem. (published online) (2020)], OrbNet predicts energies within chemical accuracy of density functional theory at a computational cost that is 1000-fold or more reduced.
Additional Information
© 2020 Published under license by AIP Publishing. Submitted: 16 July 2020; Accepted: 7 September 2020; Published Online: 25 September 2020. The authors thank Lixue Sherry Cheng for providing geometries for the DrugBank-T dataset and Anders Christensen for helpful comments on the manuscript. Z.Q. acknowledges the graduate research funding from Caltech. T.F.M. and A.A. acknowledge partial support from the Caltech DeLogi fund, and A.A. acknowledges support from a Caltech Bren professorship.Attached Files
Published - 5.0021955.pdf
Submitted - 2007.08026.pdf
Supplemental Material - drugbank-t_geometries.zip
Supplemental Material - splits.zip
Files
Additional details
- Eprint ID
- 104991
- Resolver ID
- CaltechAUTHORS:20200818-095759329
- Caltech De Logi Fund
- Bren Professor of Computing and Mathematical Sciences
- Created
-
2020-08-18Created from EPrint's datestamp field
- Updated
-
2023-05-23Created from EPrint's last_modified field