skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Representing local atomic environment using descriptors based on local correlations

Abstract

Statistical learning of material properties is an emerging topic of research and has been tremendously successful in areas such as representing complex energy landscapes as well as in technologically relevant areas, like identification of better catalysts and electronic materials. However, analysis of large data sets to efficiently learn characteristic features of a complex energy landscape, for example, depends on the ability of descriptors to effectively screen different local atomic environments. Thus, discovering appropriate descriptors of bulk or defect properties and the functional dependence of such properties on these descriptors remains a difficult and tedious process. To this end, we develop here a framework to generate descriptors based on many-body correlations that can effectively capture intrinsic geometric features of the local environment of an atom. These descriptors are based on the spectrum of two-body, three-body, four-body, and higher order correlations between an atom and its neighbors and are evaluated by calculating the corresponding two-body, three-body, and four-body overlap integrals. They are invariant to global translation, global rotation, reflection, and permutations of atomic indices. By systematically testing the ability to capture the local atomic environment, it is shown that the local correlation descriptors are able to successfully reconstruct structures containing 10-25 atomsmore » which was previously not possible.« less

Authors:
 [1]
  1. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States). Physics Division
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1502005
Report Number(s):
LLNL-JRNL-757813
Journal ID: ISSN 0021-9606; 945702
Grant/Contract Number:  
AC52-07NA27344
Resource Type:
Accepted Manuscript
Journal Name:
Journal of Chemical Physics
Additional Journal Information:
Journal Volume: 149; Journal Issue: 24; Journal ID: ISSN 0021-9606
Publisher:
American Institute of Physics (AIP)
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; 37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; Gaussian processes; interatomic potentials; graph theory; machine learning; artificial neural networks

Citation Formats

Samanta, Amit. Representing local atomic environment using descriptors based on local correlations. United States: N. p., 2018. Web. doi:10.1063/1.5055772.
Samanta, Amit. Representing local atomic environment using descriptors based on local correlations. United States. doi:10.1063/1.5055772.
Samanta, Amit. Wed . "Representing local atomic environment using descriptors based on local correlations". United States. doi:10.1063/1.5055772. https://www.osti.gov/servlets/purl/1502005.
@article{osti_1502005,
title = {Representing local atomic environment using descriptors based on local correlations},
author = {Samanta, Amit},
abstractNote = {Statistical learning of material properties is an emerging topic of research and has been tremendously successful in areas such as representing complex energy landscapes as well as in technologically relevant areas, like identification of better catalysts and electronic materials. However, analysis of large data sets to efficiently learn characteristic features of a complex energy landscape, for example, depends on the ability of descriptors to effectively screen different local atomic environments. Thus, discovering appropriate descriptors of bulk or defect properties and the functional dependence of such properties on these descriptors remains a difficult and tedious process. To this end, we develop here a framework to generate descriptors based on many-body correlations that can effectively capture intrinsic geometric features of the local environment of an atom. These descriptors are based on the spectrum of two-body, three-body, four-body, and higher order correlations between an atom and its neighbors and are evaluated by calculating the corresponding two-body, three-body, and four-body overlap integrals. They are invariant to global translation, global rotation, reflection, and permutations of atomic indices. By systematically testing the ability to capture the local atomic environment, it is shown that the local correlation descriptors are able to successfully reconstruct structures containing 10-25 atoms which was previously not possible.},
doi = {10.1063/1.5055772},
journal = {Journal of Chemical Physics},
number = 24,
volume = 149,
place = {United States},
year = {2018},
month = {12}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 1 work
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Normalized cuts and image segmentation
journal, January 2000

  • Jianbo Shi, ; Malik, J.
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, Issue 8
  • DOI: 10.1109/34.868688

A universal strategy for the creation of machine learning-based atomistic force fields
journal, September 2017


wACSF—Weighted atom-centered symmetry functions as descriptors in machine learning potentials
journal, June 2018

  • Gastegger, M.; Schwiedrzik, L.; Bittermann, M.
  • The Journal of Chemical Physics, Vol. 148, Issue 24
  • DOI: 10.1063/1.5019667

Representing potential energy surfaces by high-dimensional neural network potentials
journal, April 2014


Learning molecular energies using localized graph kernels
journal, March 2017

  • Ferré, Grégoire; Haut, Terry; Barros, Kipton
  • The Journal of Chemical Physics, Vol. 146, Issue 11
  • DOI: 10.1063/1.4978623

First-principles interatomic potentials for ten elemental metals via compressed sensing
journal, August 2015


Neural networks vs Gaussian process regression for representing potential energy surfaces: A comparative study of fit quality and vibrational spectrum accuracy
journal, June 2018

  • Kamath, Aditya; Vargas-Hernández, Rodrigo A.; Krems, Roman V.
  • The Journal of Chemical Physics, Vol. 148, Issue 24
  • DOI: 10.1063/1.5003074

Efficient nonparametric n -body force fields from machine learning
journal, May 2018


Computational aspects of many-body potentials
journal, May 2012

  • Plimpton, Steven J.; Thompson, Aidan P.
  • MRS Bulletin, Vol. 37, Issue 5
  • DOI: 10.1557/mrs.2012.96

Erratum: On representing chemical environments [Phys. Rev. B 87 , 184115 (2013)]
journal, July 2017


Using Force Matching To Determine Reactive Force Fields for Water under Extreme Thermodynamic Conditions
journal, December 2016

  • Koziol, Lucas; Fried, Laurence E.; Goldman, Nir
  • Journal of Chemical Theory and Computation, Vol. 13, Issue 1
  • DOI: 10.1021/acs.jctc.6b00707

Machine Learning Predictions of Molecular Properties: Accurate Many-Body Potentials and Nonlocality in Chemical Space
journal, June 2015

  • Hansen, Katja; Biegler, Franziska; Ramakrishnan, Raghunathan
  • The Journal of Physical Chemistry Letters, Vol. 6, Issue 12
  • DOI: 10.1021/acs.jpclett.5b00831

Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
journal, June 2003


Efficient Sampling of Atomic Configurational Spaces
journal, August 2010

  • Pártay, Lívia B.; Bartók, Albert P.; Csányi, Gábor
  • The Journal of Physical Chemistry B, Vol. 114, Issue 32
  • DOI: 10.1021/jp1012973

Exploring the free energy surface using ab initio molecular dynamics
journal, April 2016

  • Samanta, Amit; Morales, Miguel A.; Schwegler, Eric
  • The Journal of Chemical Physics, Vol. 144, Issue 16
  • DOI: 10.1063/1.4945653

A fingerprint based metric for measuring similarities of crystalline structures
journal, January 2016

  • Zhu, Li; Amsler, Maximilian; Fuhrer, Tobias
  • The Journal of Chemical Physics, Vol. 144, Issue 3
  • DOI: 10.1063/1.4940026

Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials
journal, March 2015


Assessment and Validation of Machine Learning Methods for Predicting Molecular Atomization Energies
journal, July 2013

  • Hansen, Katja; Montavon, Grégoire; Biegler, Franziska
  • Journal of Chemical Theory and Computation, Vol. 9, Issue 8
  • DOI: 10.1021/ct400195d

Adaptive machine learning framework to accelerate ab initio molecular dynamics
journal, December 2014

  • Botu, Venkatesh; Ramprasad, Rampi
  • International Journal of Quantum Chemistry, Vol. 115, Issue 16
  • DOI: 10.1002/qua.24836

Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations
journal, January 2011

  • Behler, Jörg
  • Physical Chemistry Chemical Physics, Vol. 13, Issue 40
  • DOI: 10.1039/c1cp21668f

Linearized machine-learning interatomic potentials for non-magnetic elemental metals: Limitation of pairwise descriptors and trend of predictive power
journal, June 2018

  • Takahashi, Akira; Seko, Atsuto; Tanaka, Isao
  • The Journal of Chemical Physics, Vol. 148, Issue 23
  • DOI: 10.1063/1.5027283

Geometric diffusions as a tool for harmonic analysis and structure definition of data: Diffusion maps
journal, May 2005

  • Coifman, R. R.; Lafon, S.; Lee, A. B.
  • Proceedings of the National Academy of Sciences, Vol. 102, Issue 21
  • DOI: 10.1073/pnas.0500334102

Efficient non-parametric fitting of potential energy surfaces for polyatomic molecules with Gaussian processes
journal, October 2016


Order-parameter-aided temperature-accelerated sampling for the exploration of crystal polymorphism and solid-liquid phase transitions
journal, June 2014

  • Yu, Tang-Qing; Chen, Pei-Yang; Chen, Ming
  • The Journal of Chemical Physics, Vol. 140, Issue 21
  • DOI: 10.1063/1.4878665

ChIMES: A Force Matched Potential with Explicit Three-Body Interactions for Molten Carbon
journal, November 2017

  • Lindsey, Rebecca K.; Fried, Laurence E.; Goldman, Nir
  • Journal of Chemical Theory and Computation, Vol. 13, Issue 12
  • DOI: 10.1021/acs.jctc.7b00867

Diffusion maps, spectral clustering and reaction coordinates of dynamical systems
journal, July 2006

  • Nadler, Boaz; Lafon, Stéphane; Coifman, Ronald R.
  • Applied and Computational Harmonic Analysis, Vol. 21, Issue 1, p. 113-127
  • DOI: 10.1016/j.acha.2005.07.004

Deep Potential Molecular Dynamics: A Scalable Model with the Accuracy of Quantum Mechanics
journal, April 2018


Extending the accuracy of the SNAP interatomic potential form
journal, June 2018

  • Wood, Mitchell A.; Thompson, Aidan P.
  • The Journal of Chemical Physics, Vol. 148, Issue 24
  • DOI: 10.1063/1.5017641

Sampling saddle points on a free energy surface
journal, April 2014

  • Samanta, Amit; Chen, Ming; Yu, Tang-Qing
  • The Journal of Chemical Physics, Vol. 140, Issue 16
  • DOI: 10.1063/1.4869980

Atom-centered symmetry functions for constructing high-dimensional neural network potentials
journal, February 2011

  • Behler, Jörg
  • The Journal of Chemical Physics, Vol. 134, Issue 7
  • DOI: 10.1063/1.3553717

On Unique Numbering of Atoms and Unique Codes for Molecular Graphs
journal, May 1975

  • Randic, Milan
  • Journal of Chemical Information and Modeling, Vol. 15, Issue 2
  • DOI: 10.1021/ci60002a007