English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Electronic-structure properties from atom-centered predictions of the electron density

Grisafi, A., Lewis, A., Rossi, M., & Ceriotti, M. (2022). Electronic-structure properties from atom-centered predictions of the electron density.

Item is

Files

show Files
hide Files
:
2206.14087.pdf (Preprint), 2MB
Name:
2206.14087.pdf
Description:
File downloaded from arXiv at 2022-07-14
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
2022
Copyright Info:
© the Author(s)

Locators

show
hide
Locator:
https://arxiv.org/abs/2206.14087 (Preprint)
Description:
-
OA-Status:

Creators

show
hide
 Creators:
Grisafi, A.1, Author
Lewis, A.2, Author           
Rossi, M.2, Author           
Ceriotti, M.3, Author
Affiliations:
1PASTEUR, Département de Chimie,École Normale Supérieure, ou_persistent22              
2Simulations from Ab Initio Approaches, Theory Department, Max Planck Institute for the Structure and Dynamics of Matter, Max Planck Society, ou_3185035              
3Laboratory of Computational Science and Modeling, IMX, École Polytechnique Fédérale de Lausanne, ou_persistent22              

Content

show
hide
Free keywords: Physics, Chemical Physics, physics.chem-ph, Condensed Matter, Materials Science, cond-mat.mtrl-sci,Statistics, Machine Learning, stat.ML
 Abstract: The electron density of a molecule or material has recently received major attention as a target quantity of machine-learning models. A natural choice to construct a model that yields transferable and linear-scaling predictions is to represent the scalar field using a multi-centered atomic basis analogous to that routinely used in density fitting approximations. However, the non-orthogonality of the basis poses challenges for the learning exercise, as it requires accounting for all the atomic density components at once. We devise a gradient-based approach to directly minimize the loss function of the regression problem in an optimized and highly sparse feature space. In so doing, we overcome the limitations associated with adopting an atom-centered model to learn the electron density over arbitrarily complex datasets, obtaining extremely accurate predictions. The enhanced framework is tested on 32-molecule periodic cells of liquid water, presenting enough complexity to require an optimal balance between accuracy and computational efficiency. We show that starting from the predicted density a single Kohn-Sham diagonalization step can be performed to access total energy components that carry an error of just 0.1 meV/atom with respect to the reference density functional calculations. Finally, we test our method on the highly heterogeneous QM9 benchmark dataset, showing that a small fraction of the training data is enough to derive ground-state total energies within chemical accuracy.

Details

show
hide
Language(s): eng - English
 Dates: 2022-06-28
 Publication Status: Published online
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: No review
 Identifiers: arXiv: 2206.14087
 Degree: -

Event

show

Legal Case

show

Project information

show

Source

show