English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Learning Electron Densities in the Condensed Phase

Lewis, A., Grisafi, A., Ceriotti, M., & Rossi, M. (2021). Learning Electron Densities in the Condensed Phase. Journal of Chemical Theory and Computation, 17(11), 7203-7214. doi:10.1021/acs.jctc.1c00576.

Item is

Basic

show hide
Genre: Journal Article

Files

show Files
hide Files
:
ct1c00576_si_002.pdf (Supplementary material), 387KB
Name:
ct1c00576_si_002.pdf
Description:
Supporting Information: Equations for the calculation of the overlap matrix and vector of projections in periodic systems, an error analysis of the electrostatic and Hartree energies, the optimizations of the SALTED hyper-paramaters for the homogeneous and heterogeneous data sets, the optimization of the direct GPR hyper-parameters and their learning curves, and an illustration of the application of SALTED to isolated molecules using NAOs
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-
License:
-
:
acs.jctc.1c00576.pdf (Publisher version), 2MB
Name:
acs.jctc.1c00576.pdf
Description:
Open access funded by Max Planck Society.
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
2021
Copyright Info:
© The Authors. Published byAmerican Chemical Society

Locators

show
hide
Locator:
https://arxiv.org/abs/2106.05364 (Preprint)
Description:
-
OA-Status:
Locator:
https://dx.doi.org/10.1021/acs.jctc.1c00576 (Publisher version)
Description:
-
OA-Status:

Creators

show
hide
 Creators:
Lewis, A.1, Author           
Grisafi, A.2, Author
Ceriotti, M.2, Author
Rossi, M.1, Author           
Affiliations:
1Simulations from Ab Initio Approaches, Theory Department, Max Planck Institute for the Structure and Dynamics of Matter, Max Planck Society, ou_3185035              
2Laboratory of Computational Science and Modeling, IMX, École Polytechnique Féd́erale de Lausanne, ou_persistent22              

Content

show
hide
Free keywords: -
 Abstract: We introduce a local machine-learning method for predicting the electron densities of periodic systems. The framework is based on a numerical, atom-centered auxiliary basis, which enables an accurate expansion of the all-electron density in a form suitable for learning isolated and periodic systems alike. We show that, using this formulation, the electron densities of metals, semiconductors, and molecular crystals can all be accurately predicted using symmetry-adapted Gaussian process regression models, properly adjusted for the nonorthogonal nature of the basis. These predicted densities enable the efficient calculation of electronic properties, which present errors on the order of tens of meV/atom when compared to ab initio density-functional calculations. We demonstrate the key power of this approach by using a model trained on ice unit cells containing only 4 water molecules to predict the electron densities of cells containing up to 512 molecules and see no increase in the magnitude of the errors of derived electronic properties when increasing the system size. Indeed, we find that these extrapolated derived energies are more accurate than those predicted using a direct machine-learning model. Finally, on heterogeneous data sets SALTED can predict electron densities with errors below 4%.

Details

show
hide
Language(s): eng - English
 Dates: 2021-06-102021-10-202021-11-09
 Publication Status: Published in print
 Pages: 12
 Publishing info: -
 Table of Contents: -
 Rev. Type: Peer
 Identifiers: arXiv: 2106.05364
DOI: 10.1021/acs.jctc.1c00576
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: Journal of Chemical Theory and Computation
  Other : J. Chem. Theory Comput.
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: Washington, D.C. : American Chemical Society
Pages: - Volume / Issue: 17 (11) Sequence Number: - Start / End Page: 7203 - 7214 Identifier: ISSN: 1549-9618
CoNE: https://pure.mpg.de/cone/journals/resource/111088195283832