Efficient and accurate machine-learning interpolation of atomic energies in compositions with many species

Nongnuch Artrith, Alexander Urban, and Gerbrand Ceder
Phys. Rev. B 96, 014112 – Published 21 July 2017
PDFHTMLExport Citation

Abstract

Machine-learning potentials (MLPs) for atomistic simulations are a promising alternative to conventional classical potentials. Current approaches rely on descriptors of the local atomic environment with dimensions that increase quadratically with the number of chemical species. In this paper, we demonstrate that such a scaling can be avoided in practice. We show that a mathematically simple and computationally efficient descriptor with constant complexity is sufficient to represent transition-metal oxide compositions and biomolecules containing 11 chemical species with a precision of around 3 meV/atom. This insight removes a perceived bound on the utility of MLPs and paves the way to investigate the physics of previously inaccessible materials with more than ten chemical species.

  • Figure
  • Figure
  • Figure
  • Received 5 May 2017
  • Revised 20 June 2017

DOI:https://doi.org/10.1103/PhysRevB.96.014112

Published by the American Physical Society

Physics Subject Headings (PhySH)

Condensed Matter, Materials & Applied PhysicsInterdisciplinary Physics

Authors & Affiliations

Nongnuch Artrith1,*, Alexander Urban1, and Gerbrand Ceder1,2,†

  • 1Department of Materials Science and Engineering, University of California, Berkeley, California 94720, USA
  • 2Materials Science Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA

  • *nartrith@berkeley.edu
  • gceder@berkeley.edu

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand
Issue

Vol. 96, Iss. 1 — 1 July 2017

Reuse & Permissions
Access Options
CHORUS

Article Available via CHORUS

Download Accepted Manuscript
Author publication services for translation and copyediting assistance advertisement

Authorization Required


×
×

Images

×

Sign up to receive regular email alerts from Physical Review B

Log In

Cancel
×

Search


Article Lookup

Paste a citation or DOI

Enter a citation
×