We propose a grid-based local representation of electronic quantities that can be used in machine learning applications for molecules, which is compact, fixed in size, and able to distinguish different chemical environments. We apply the proposed approach to represent the external potential in density functional theory with modified pseudopotentials and demonstrate its proof of concept by predicting the Perdew-Burke-Ernzerhof and local density approximation electronic density and exchange-correlation potentials by kernel ridge regression. For 16 small molecules consisting of C, H, N, and O, the mean absolute error of exchange-correlation energy was 0.78 kcal/mol when trained for individual molecules. Furthermore, the model is shown to predict the exchange-correlation energy with an accuracy of 3.68 kcal/mol when the model is trained with a small fraction (4%) of all 16 molecules of the present dataset, suggesting a promising possibility that the current machine-learned model may predict the exchange-correlation energies of an arbitrary molecule with reasonable accuracy when trained with a sufficient amount of data covering an extensive variety of chemical environments.
Skip Nav Destination
Article navigation
28 June 2018
Research Article|
June 22 2018
A local environment descriptor for machine-learned density functional theory at the generalized gradient approximation level
Special Collection:
Data-Enabled Theoretical Chemistry
Hyunjun Ji;
Hyunjun Ji
1
Graduate School of EEWS, KAIST
, Daejeon, South Korea
Search for other works by this author on:
Yousung Jung
Yousung Jung
a)
1
Graduate School of EEWS, KAIST
, Daejeon, South Korea
2
Department of Chemical and Biomolecular Engineering, KAIST
, Daejeon, South Korea
Search for other works by this author on:
a)
Email: ysjn@kaist.ac.kr.
J. Chem. Phys. 148, 241742 (2018)
Article history
Received:
January 19 2018
Accepted:
June 01 2018
Citation
Hyunjun Ji, Yousung Jung; A local environment descriptor for machine-learned density functional theory at the generalized gradient approximation level. J. Chem. Phys. 28 June 2018; 148 (24): 241742. https://doi.org/10.1063/1.5022839
Download citation file:
Pay-Per-View Access
$40.00
Sign In
You could not be signed in. Please check your credentials and make sure you have an active account and try again.
Citing articles via
DeePMD-kit v2: A software package for deep potential models
Jinzhe Zeng, Duo Zhang, et al.
CREST—A program for the exploration of low-energy molecular chemical space
Philipp Pracht, Stefan Grimme, et al.
Related Content
Compositional descriptor-based recommender system for the materials discovery
J. Chem. Phys. (March 2018)
wACSF—Weighted atom-centered symmetry functions as descriptors in machine learning potentials
J. Chem. Phys. (March 2018)
Improve the performance of machine-learning potentials by optimizing descriptors
J. Chem. Phys. (June 2019)
Descriptors for predicting the lattice constant of body centered cubic crystal
J. Chem. Phys. (May 2017)
Genarris: Random generation of molecular crystal structures and fast screening with a Harris approximation
J. Chem. Phys. (March 2018)