Discovering charge density functionals and structure-property relationships with PROPhet: A general framework for coupling machine learning and first-principles methods
Abstract
Modern ab initio methods have rapidly increased our understanding of solid state materials properties, chemical reactions, and the quantum interactions between atoms. However, poor scaling often renders direct ab initio calculations intractable for large or complex systems. There are two obvious avenues through which to remedy this problem: (i) develop new, less expensive methods to calculate system properties, or (ii) make existing methods faster. This paper describes an open source framework designed to pursue both of these avenues. PROPhet (short for PROPerty Prophet) utilizes machine learning techniques to find complex, non-linear mappings between sets of material or system properties. The result is a single code capable of learning analytical potentials, non-linear density functionals, and other structure-property or property-property relationships. These capabilities enable highly accurate mesoscopic simulations, facilitate computation of expensive properties, and enable the development of predictive models for systematic materials design and optimization. Here, this work explores the coupling of machine learning to ab initio methods through means both familiar (e.g., the creation of various potentials and energy functionals) and less familiar (e.g., the creation of density functionals for arbitrary properties), serving both to demonstrate PROPhet’s ability to create exciting post-processing analysis tools and to open the door tomore »
- Authors:
- Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States). Dept. of Mechanical Engineering; Univ. of New Mexico, Albuquerque, NM (United States). Dept. of Chemistry and Chemical Biology
- Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States). Dept. of Mechanical Engineering
- Publication Date:
- Research Org.:
- Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States); Alliance for Sustainable Energy, LLC, Golden, CO (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Basic Energy Sciences (BES) (SC-22)USDOE Office of Energy Efficiency and Renewable Energy (EERE)
- OSTI Identifier:
- 1368375
- Grant/Contract Number:
- SC0001299; FG02-09ER46577; AC36-8GO28308
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Scientific Reports
- Additional Journal Information:
- Journal Volume: 7; Journal Issue: 1; Journal ID: ISSN 2045-2322
- Publisher:
- Nature Publishing Group
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING
Citation Formats
Kolb, Brian, Lentz, Levi C., and Kolpak, Alexie M. Discovering charge density functionals and structure-property relationships with PROPhet: A general framework for coupling machine learning and first-principles methods. United States: N. p., 2017.
Web. doi:10.1038/s41598-017-01251-z.
Kolb, Brian, Lentz, Levi C., & Kolpak, Alexie M. Discovering charge density functionals and structure-property relationships with PROPhet: A general framework for coupling machine learning and first-principles methods. United States. doi:10.1038/s41598-017-01251-z.
Kolb, Brian, Lentz, Levi C., and Kolpak, Alexie M. Wed .
"Discovering charge density functionals and structure-property relationships with PROPhet: A general framework for coupling machine learning and first-principles methods". United States. doi:10.1038/s41598-017-01251-z. https://www.osti.gov/servlets/purl/1368375.
@article{osti_1368375,
title = {Discovering charge density functionals and structure-property relationships with PROPhet: A general framework for coupling machine learning and first-principles methods},
author = {Kolb, Brian and Lentz, Levi C. and Kolpak, Alexie M.},
abstractNote = {Modern ab initio methods have rapidly increased our understanding of solid state materials properties, chemical reactions, and the quantum interactions between atoms. However, poor scaling often renders direct ab initio calculations intractable for large or complex systems. There are two obvious avenues through which to remedy this problem: (i) develop new, less expensive methods to calculate system properties, or (ii) make existing methods faster. This paper describes an open source framework designed to pursue both of these avenues. PROPhet (short for PROPerty Prophet) utilizes machine learning techniques to find complex, non-linear mappings between sets of material or system properties. The result is a single code capable of learning analytical potentials, non-linear density functionals, and other structure-property or property-property relationships. These capabilities enable highly accurate mesoscopic simulations, facilitate computation of expensive properties, and enable the development of predictive models for systematic materials design and optimization. Here, this work explores the coupling of machine learning to ab initio methods through means both familiar (e.g., the creation of various potentials and energy functionals) and less familiar (e.g., the creation of density functionals for arbitrary properties), serving both to demonstrate PROPhet’s ability to create exciting post-processing analysis tools and to open the door to improving ab initio methods themselves with these powerful machine learning techniques.},
doi = {10.1038/s41598-017-01251-z},
journal = {Scientific Reports},
number = 1,
volume = 7,
place = {United States},
year = {2017},
month = {4}
}
Web of Science
Figures / Tables:

Works referenced in this record:
Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
journal, January 2012
- Rupp, Matthias; Tkatchenko, Alexandre; Müller, Klaus-Robert
- Physical Review Letters, Vol. 108, Issue 5
Density‐functional thermochemistry. III. The role of exact exchange
journal, April 1993
- Becke, Axel D.
- The Journal of Chemical Physics, Vol. 98, Issue 7, p. 5648-5652
Finding Density Functionals with Machine Learning
journal, June 2012
- Snyder, John C.; Rupp, Matthias; Hansen, Katja
- Physical Review Letters, Vol. 108, Issue 25
Permutation invariant polynomial neural network approach to fitting potential energy surfaces
journal, August 2013
- Jiang, Bin; Guo, Hua
- The Journal of Chemical Physics, Vol. 139, Issue 5
Kinetic-energy functional of the electron density
journal, June 1992
- Wang, Lin-Wang; Teter, Michael P.
- Physical Review B, Vol. 45, Issue 23
Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set
journal, July 1996
- Kresse, G.; Furthmüller, J.
- Computational Materials Science, Vol. 6, Issue 1, p. 15-50
Representing potential energy surfaces by high-dimensional neural network potentials
journal, April 2014
- Behler, J.
- Journal of Physics: Condensed Matter, Vol. 26, Issue 18
Time-dependent density functional theory: Past, present, and future
journal, August 2005
- Burke, Kieron; Werschnik, Jan; Gross, E. K. U.
- The Journal of Chemical Physics, Vol. 123, Issue 6
Understanding the Composition and Activity of Electrocatalytic Nanoalloys in Aqueous Solvents: A Combination of DFT and Accurate Neural Network Potentials
journal, April 2014
- Artrith, Nongnuch; Kolpak, Alexie M.
- Nano Letters, Vol. 14, Issue 5
Ab initio calculations of optical absorption spectra: Solution of the Bethe–Salpeter equation within density matrix perturbation theory
journal, October 2010
- Rocca, Dario; Lu, Deyu; Galli, Giulia
- The Journal of Chemical Physics, Vol. 133, Issue 16
QUANTUM ESPRESSO: a modular and open-source software project for quantum simulations of materials
journal, September 2009
- Giannozzi, Paolo; Baroni, Stefano; Bonini, Nicola
- Journal of Physics: Condensed Matter, Vol. 21, Issue 39, Article No. 395502
Lattice Dynamics of Diamond
journal, June 1967
- Warren, J. L.; Yarnell, J. L.; Dolling, G.
- Physical Review, Vol. 158, Issue 3
Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set
journal, October 1996
- Kresse, G.; Furthmüller, J.
- Physical Review B, Vol. 54, Issue 16, p. 11169-11186
First principles phonon calculations in materials science
journal, November 2015
- Togo, Atsushi; Tanaka, Isao
- Scripta Materialia, Vol. 108
Resolution-of-identity approach to Hartree–Fock, hybrid density functionals, RPA, MP2 and GW with numeric atom-centered orbital basis functions
journal, May 2012
- Ren, Xinguo; Rinke, Patrick; Blum, Volker
- New Journal of Physics, Vol. 14, Issue 5
Potential Energy Surfaces Fitted by Artificial Neural Networks
journal, March 2010
- Handley, Chris M.; Popelier, Paul L. A.
- The Journal of Physical Chemistry A, Vol. 114, Issue 10
The GW method
journal, March 1998
- Aryasetiawan, F.; Gunnarsson, O.
- Reports on Progress in Physics, Vol. 61, Issue 3
Fast Parallel Algorithms for Short-Range Molecular Dynamics
journal, March 1995
- Plimpton, Steve
- Journal of Computational Physics, Vol. 117, Issue 1
Inhomogeneous Electron Gas
journal, November 1964
- Hohenberg, P.; Kohn, W.
- Physical Review, Vol. 136, Issue 3B, p. B864-B871
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces
journal, April 2007
- Behler, Jörg; Parrinello, Michele
- Physical Review Letters, Vol. 98, Issue 14
Orbital-free bond breaking via machine learning
journal, December 2013
- Snyder, John C.; Rupp, Matthias; Hansen, Katja
- The Journal of Chemical Physics, Vol. 139, Issue 22
Machine learning of molecular electronic properties in chemical compound space
journal, September 2013
- Montavon, Grégoire; Rupp, Matthias; Gobre, Vivekanand
- New Journal of Physics, Vol. 15, Issue 9
Thermal conductivity of group-IV semiconductors from a kinetic-collective model
journal, September 2014
- de Tomas, C.; Cantarero, A.; Lopeandia, A. F.
- Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 470, Issue 2169
Ab initio molecular simulations with numeric atom-centered orbitals
journal, November 2009
- Blum, Volker; Gehrke, Ralf; Hanke, Felix
- Computer Physics Communications, Vol. 180, Issue 11
Representing high-dimensional potential-energy surfaces for reactions at surfaces by neural networks
journal, September 2004
- Lorenz, Sönke; Groß, Axel; Scheffler, Matthias
- Chemical Physics Letters, Vol. 395, Issue 4-6
Modified Statistical Treatment of Kinetic Energy in the Thomas−Fermi Model †
journal, May 2004
- Chai, Jeng-Da; Weeks, John D.
- The Journal of Physical Chemistry B, Vol. 108, Issue 21
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
Neural network approach to quantum-chemistry data: Accurate prediction of density functional theory energies
journal, August 2009
- Balabin, Roman M.; Lomakina, Ekaterina I.
- The Journal of Chemical Physics, Vol. 131, Issue 7
Grand canonical molecular dynamics simulations of Cu–Au nanoalloys in thermal equilibrium using reactive ANN potentials
journal, December 2015
- Artrith, Nongnuch; Kolpak, Alexie M.
- Computational Materials Science, Vol. 110
Ab initiomolecular dynamics for liquid metals
journal, January 1993
- Kresse, G.; Hafner, J.
- Physical Review B, Vol. 47, Issue 1, p. 558-561
Neural network molecular dynamics simulations of solid–liquid interfaces: water at low-index copper surfaces
journal, January 2016
- Natarajan, Suresh Kondati; Behler, Jörg
- Physical Chemistry Chemical Physics, Vol. 18, Issue 41
Kinetic Energy of Hydrocarbons as a Function of Electron Density and Convolutional Neural Networks
journal, February 2016
- Yao, Kun; Parkhill, John
- Journal of Chemical Theory and Computation, Vol. 12, Issue 3
Picosecond optical studies of amorphous diamond and diamondlike carbon: Thermal conductivity and longitudinal sound velocity
journal, September 1994
- Morath, Christopher J.; Maris, Humphrey J.; Cuomo, Jerome J.
- Journal of Applied Physics, Vol. 76, Issue 5
Pure density functional for strong correlation and the thermodynamic limit from machine learning
journal, December 2016
- Li, Li; Baker, Thomas E.; White, Steven R.
- Physical Review B, Vol. 94, Issue 24
NWChem: A comprehensive and scalable open-source solution for large scale molecular simulations
journal, September 2010
- Valiev, M.; Bylaska, E. J.; Govind, N.
- Computer Physics Communications, Vol. 181, Issue 9, p. 1477-1489
Coupled-cluster theory in quantum chemistry
journal, February 2007
- Bartlett, Rodney J.; Musiał, Monika
- Reviews of Modern Physics, Vol. 79, Issue 1
Self-Consistent Equations Including Exchange and Correlation Effects
journal, November 1965
- Kohn, W.; Sham, L. J.
- Physical Review, Vol. 140, Issue 4A, p. A1133-A1138
A general-purpose machine learning framework for predicting properties of inorganic materials
journal, August 2016
- Ward, Logan; Agrawal, Ankit; Choudhary, Alok
- npj Computational Materials, Vol. 2, Issue 1
On the limited memory BFGS method for large scale optimization
journal, August 1989
- Liu, Dong C.; Nocedal, Jorge
- Mathematical Programming, Vol. 45, Issue 1-3
Understanding machine-learned density functionals: Understanding Machine-Learned Density Functionals
journal, November 2015
- Li, Li; Snyder, John C.; Pelaschier, Isabelle M.
- International Journal of Quantum Chemistry, Vol. 116, Issue 11
Figures / Tables found in this record: