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

Title: Genetic algorithms for computational materials discovery accelerated by machine learning

Abstract

Materials discovery is increasingly being impelled by machine learning methods that rely on pre-existing datasets. Where datasets are lacking, unbiased data generation can be achieved with genetic algorithms. Here a machine learning model is trained on-the-fly as a computationally inexpensive energy predictor before analyzing how to augment convergence in genetic algorithm-based approaches by using the model as a surrogate. This leads to a machine learning accelerated genetic algorithm combining robust qualities of the genetic algorithm with rapid machine learning. The approach is used to search for stable, compositionally variant, geometrically similar nanoparticle alloys to illustrate its capability for accelerated materials discovery, e.g., nanoalloy catalysts. The machine learning accelerated approach, in this case, yields a 50-fold reduction in the number of required energy calculations compared to a traditional “brute force” genetic algorithm. This makes searching through the space of all homotops and compositions of a binary alloy particle in a given structure feasible, using density functional theory calculations.

Authors:
 [1]; ORCiD logo [2];  [3]; ORCiD logo [2];  [1]
  1. Stanford Univ., Stanford, CA (United States); SLAC National Accelerator Lab., Menlo Park, CA (United States)
  2. Technical Univ. of Denmark, Lyngby (Denmark)
  3. Toyota Research Institute, Los Altos, CA (United States)
Publication Date:
Research Org.:
SLAC National Accelerator Lab., Menlo Park, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1511649
Grant/Contract Number:  
AC02-76SF0051
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
npj Computational Materials
Additional Journal Information:
Journal Volume: 5; Journal Issue: 1; Journal ID: ISSN 2057-3960
Publisher:
Nature Publishing Group
Country of Publication:
United States
Language:
English
Subject:
36 MATERIALS SCIENCE

Citation Formats

Jennings, Paul C., Lysgaard, Steen, Hummelshøj, Jens Strabo, Vegge, Tejs, and Bligaard, Thomas. Genetic algorithms for computational materials discovery accelerated by machine learning. United States: N. p., 2019. Web. doi:10.1038/s41524-019-0181-4.
Jennings, Paul C., Lysgaard, Steen, Hummelshøj, Jens Strabo, Vegge, Tejs, & Bligaard, Thomas. Genetic algorithms for computational materials discovery accelerated by machine learning. United States. doi:10.1038/s41524-019-0181-4.
Jennings, Paul C., Lysgaard, Steen, Hummelshøj, Jens Strabo, Vegge, Tejs, and Bligaard, Thomas. Wed . "Genetic algorithms for computational materials discovery accelerated by machine learning". United States. doi:10.1038/s41524-019-0181-4. https://www.osti.gov/servlets/purl/1511649.
@article{osti_1511649,
title = {Genetic algorithms for computational materials discovery accelerated by machine learning},
author = {Jennings, Paul C. and Lysgaard, Steen and Hummelshøj, Jens Strabo and Vegge, Tejs and Bligaard, Thomas},
abstractNote = {Materials discovery is increasingly being impelled by machine learning methods that rely on pre-existing datasets. Where datasets are lacking, unbiased data generation can be achieved with genetic algorithms. Here a machine learning model is trained on-the-fly as a computationally inexpensive energy predictor before analyzing how to augment convergence in genetic algorithm-based approaches by using the model as a surrogate. This leads to a machine learning accelerated genetic algorithm combining robust qualities of the genetic algorithm with rapid machine learning. The approach is used to search for stable, compositionally variant, geometrically similar nanoparticle alloys to illustrate its capability for accelerated materials discovery, e.g., nanoalloy catalysts. The machine learning accelerated approach, in this case, yields a 50-fold reduction in the number of required energy calculations compared to a traditional “brute force” genetic algorithm. This makes searching through the space of all homotops and compositions of a binary alloy particle in a given structure feasible, using density functional theory calculations.},
doi = {10.1038/s41524-019-0181-4},
journal = {npj Computational Materials},
issn = {2057-3960},
number = 1,
volume = 5,
place = {United States},
year = {2019},
month = {4}
}

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

Save / Share:

Works referenced in this record:

Octahedral PtNi Nanoparticle Catalysts Exceptional Oxygen Reduction Activity by Tuning the Alloy Particle Surface Composition
journal, October 2012

  • Cui, Chunhua; Gan, Lin; Li, Hui-Hui
  • Nano Letters, Vol. 12, Issue 11, p. 5885-5889
  • DOI: 10.1021/nl3032795

Nanoalloys:� From Theory to Applications of Alloy Clusters and Nanoparticles
journal, March 2008

  • Ferrando, Riccardo; Jellinek, Julius; Johnston, Roy L.
  • Chemical Reviews, Vol. 108, Issue 3, p. 845-910
  • DOI: 10.1021/cr040090g

Systematic Study of Au 6 to Au 12 Gold Clusters on MgO(100) F Centers Using Density-Functional Theory
journal, March 2012