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

Title: Fast and Accurate Predictions of Total Energy for Solid Solution Alloys with Graph Convolutional Neural Networks

Abstract

We use graph convolutional neural networks (GCNNs) to produce fast and accurate predictions of the total energy of solid solution binary alloys. GCNNs allow us to abstract the lattice structure of a solid material as a graph, whereby atoms are modeled as nodes and metallic bonds as edges. This representation naturally incorporates information about the structure of the material, thereby eliminating the need for computationally expensive data pre-processing which would be required with standard neural network (NN) approaches. We train GCNNs on ab-initio density functional theory (DFT) for copper-gold (CuAu) and iron-platinum (FePt) data that has been generated by running the LSMS-3 code, which implements a locally self-consistent multiple scattering method, on OLCF supercomputers Titan and Summit. GCNN outperforms the ab-initio DFT simulation by orders of magnitude in terms of computational time to produce the estimate of the total energy for a given atomic configuration of the lattice structure. We compare the predictive performance of GCNN models against a standard NN such as dense feedforward multi-layer perceptron (MLP) by using the root-mean-squared errors to quantify the predictive quality of the deep learning (DL) models. We find that the attainable accuracy of GCNNs is at least an order of magnitude bettermore » than that of the MLP.« less

Authors:
ORCiD logo [1];  [2];  [1]; ORCiD logo [1]; ORCiD logo [2]
  1. ORNL
  2. Polytechnic University of Milan (Politecnico di Milano), Italy
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1855626
DOE Contract Number:  
AC05-00OR22725
Resource Type:
Conference
Resource Relation:
Journal Volume: 1512; Conference: Smoky Mountains Computational Sciences and Engineering (SMC 2021) - Virtual, Tennessee, United States of America - 10/18/2021 4:00:00 AM-10/20/2021 4:00:00 AM
Country of Publication:
United States
Language:
English

Citation Formats

Lupo Pasini, Massimiliano, Burcul, Marco, Reeve, Sam, Eisenbach, Markus, and Perotto, Simona. Fast and Accurate Predictions of Total Energy for Solid Solution Alloys with Graph Convolutional Neural Networks. United States: N. p., 2022. Web. doi:10.1007/978-3-030-96498-6_5.
Lupo Pasini, Massimiliano, Burcul, Marco, Reeve, Sam, Eisenbach, Markus, & Perotto, Simona. Fast and Accurate Predictions of Total Energy for Solid Solution Alloys with Graph Convolutional Neural Networks. United States. https://doi.org/10.1007/978-3-030-96498-6_5
Lupo Pasini, Massimiliano, Burcul, Marco, Reeve, Sam, Eisenbach, Markus, and Perotto, Simona. 2022. "Fast and Accurate Predictions of Total Energy for Solid Solution Alloys with Graph Convolutional Neural Networks". United States. https://doi.org/10.1007/978-3-030-96498-6_5. https://www.osti.gov/servlets/purl/1855626.
@article{osti_1855626,
title = {Fast and Accurate Predictions of Total Energy for Solid Solution Alloys with Graph Convolutional Neural Networks},
author = {Lupo Pasini, Massimiliano and Burcul, Marco and Reeve, Sam and Eisenbach, Markus and Perotto, Simona},
abstractNote = {We use graph convolutional neural networks (GCNNs) to produce fast and accurate predictions of the total energy of solid solution binary alloys. GCNNs allow us to abstract the lattice structure of a solid material as a graph, whereby atoms are modeled as nodes and metallic bonds as edges. This representation naturally incorporates information about the structure of the material, thereby eliminating the need for computationally expensive data pre-processing which would be required with standard neural network (NN) approaches. We train GCNNs on ab-initio density functional theory (DFT) for copper-gold (CuAu) and iron-platinum (FePt) data that has been generated by running the LSMS-3 code, which implements a locally self-consistent multiple scattering method, on OLCF supercomputers Titan and Summit. GCNN outperforms the ab-initio DFT simulation by orders of magnitude in terms of computational time to produce the estimate of the total energy for a given atomic configuration of the lattice structure. We compare the predictive performance of GCNN models against a standard NN such as dense feedforward multi-layer perceptron (MLP) by using the root-mean-squared errors to quantify the predictive quality of the deep learning (DL) models. We find that the attainable accuracy of GCNNs is at least an order of magnitude better than that of the MLP.},
doi = {10.1007/978-3-030-96498-6_5},
url = {https://www.osti.gov/biblio/1855626}, journal = {},
issn = {1865--0929},
number = ,
volume = 1512,
place = {United States},
year = {Tue Mar 01 00:00:00 EST 2022},
month = {Tue Mar 01 00:00:00 EST 2022}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: