Incorporating long-range physics in atomic-scale machine learning

JSON Export

{
  "id": "284", 
  "updated": "2019-12-18T00:00:00+00:00", 
  "metadata": {
    "version": 1, 
    "contributors": [
      {
        "givennames": "Andrea", 
        "affiliations": [
          "Laboratory of Computational Science and Modeling, Institute of Materials, E\u0301cole Polytechnique Fe\u0301de\u0301rale de Lausanne, 1015 Lausanne, Switzerland"
        ], 
        "familyname": "Grisafi"
      }, 
      {
        "givennames": "Michele", 
        "affiliations": [
          "Laboratory of Computational Science and Modeling, Institute of Materials, E\u0301cole Polytechnique Fe\u0301de\u0301rale de Lausanne, 1015 Lausanne, Switzerland"
        ], 
        "email": "michele.ceriotti@epfl.ch", 
        "familyname": "Ceriotti"
      }
    ], 
    "title": "Incorporating long-range physics in atomic-scale machine learning", 
    "_oai": {
      "id": "oai:materialscloud.org:284"
    }, 
    "keywords": [
      "ERC", 
      "MARVEL/DD1", 
      "machine learning", 
      "long-range interactions"
    ], 
    "publication_date": "Dec 18, 2019, 00:00:00", 
    "_files": [
      {
        "key": "binding_energies.dat", 
        "description": "Binding energies of molecular dimers.", 
        "checksum": "md5:6421bec70027a9f349f868fa5d7b8019", 
        "size": 253776
      }, 
      {
        "key": "README.txt", 
        "description": "README file.", 
        "checksum": "md5:fb7ee4cbfb6cdd75cbdb7b632850a2a3", 
        "size": 1161
      }, 
      {
        "key": "charged_dimers_binding_with_monomers.xyz", 
        "description": "Atomic coordinates of molecular dimers.", 
        "checksum": "md5:e38dfea4ccc41ea1dc400eb58ca56d3a", 
        "size": 7276575
      }
    ], 
    "references": [
      {
        "comment": "Paper in which the method is described", 
        "doi": "10.1063/1.5128375", 
        "citation": "A. Grisafi, M. Ceriotti, J. Chem. Phys. 151, 204105 (2019)", 
        "url": "https://aip.scitation.org/doi/10.1063/1.5128375", 
        "type": "Journal reference"
      }, 
      {
        "comment": "Preprint where the method is described.", 
        "doi": "", 
        "citation": "A. Grisafi, M. Ceriotti, arXiv:1909.04512", 
        "url": "https://arxiv.org/abs/1909.04512", 
        "type": "Preprint"
      }
    ], 
    "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.", 
    "status": "published", 
    "license": "Creative Commons Attribution 4.0 International", 
    "conceptrecid": "283", 
    "is_last": true, 
    "mcid": "2019.0090/v1", 
    "edited_by": 98, 
    "id": "284", 
    "owner": 23, 
    "license_addendum": "", 
    "doi": "10.24435/materialscloud:2019.0090/v1"
  }, 
  "revision": 1, 
  "created": "2020-05-12T13:53:33.941222+00:00"
}