materialscloud:2019.0090/v1

Incorporating long-range physics in atomic-scale machine learning

Andrea Grisafi1, Michele Ceriotti1*

1 Laboratory of Computational Science and Modeling, Institute of Materials, École Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland

* Corresponding authors emails: michele.ceriotti@epfl.ch
DOI10.24435/materialscloud:2019.0090/v1 [version v1]

Publication date: Dec 18, 2019

How to cite this record

Andrea Grisafi, Michele Ceriotti, Incorporating long-range physics in atomic-scale machine learning, Materials Cloud Archive 2019.0090/v1 (2019), doi: 10.24435/materialscloud:2019.0090/v1.

Description

The most successful and popular machine learning models of atomic-scale properties derive their transferability from a locality ansatz. The properties of a large molecule or a bulk material are written as a sum over contributions that depend on the configurations within finite atom-centered environments. The obvious downside of this approach is that it cannot capture nonlocal, nonadditive effects such as those arising due to long-range electrostatics or quantum interference. We propose a solution to this problem by introducing nonlocal representations of the system, which are remapped as feature vectors that are defined locally and are equivariant in O(3). We consider, in particular, one form that has the same asymptotic behavior as the electrostatic potential. We demonstrate that this framework can capture nonlocal, long-range physics by building a model for the electrostatic energy of randomly distributed point-charges, for the unrelaxed binding curves of charged organic molecular dimers, and for the electronic dielectric response of liquid water. By combining a representation of the system that is sensitive to long-range correlations with the transferability of an atom-centered additive model, this method outperforms current state-of-the-art machine-learning schemes and provides a conceptual framework to incorporate nonlocal physics into atomistic machine learning.

Materials Cloud sections using this data

No Explore or Discover sections associated with this archive record.

Files

File name Size Description
binding_energies.dat
MD5md5:6421bec70027a9f349f868fa5d7b8019
247.8 KiB Binding energies of molecular dimers.
README.txt
MD5md5:fb7ee4cbfb6cdd75cbdb7b632850a2a3
1.1 KiB README file.
charged_dimers_binding_with_monomers.xyz
MD5md5:e38dfea4ccc41ea1dc400eb58ca56d3a
6.9 MiB Atomic coordinates of molecular dimers.

License

Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.

External references

Journal reference (Paper in which the method is described)
Preprint (Preprint where the method is described.)

Keywords

ERC MARVEL/DD1 machine learning long-range interactions

Version history:

2019.0090/v1 (version v1) [This version] Dec 18, 2019 DOI10.24435/materialscloud:2019.0090/v1