Comparison of permutationally invariant polynomials, neural networks, and Gaussian approximation potentials in representing water interactions through many-body expansions
Top Cited Papers
- 9 April 2018
- journal article
- conference paper
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 148 (24) , 241725
Abstract
The accurate representation of multidimensional potential energy surfaces is a necessary requirement for realistic computer simulations of molecular systems. The continued increase in computer power accompanied by advances in correlated electronic structure methods nowadays enables routine calculations of accurate interaction energies for small systems, which can then be used as references for the development of analytical potential energy functions (PEFs) rigorously derived from many-body (MB) expansions. Building on the accuracy of the MB-pol many-body PEF, we investigate here the performance of permutationally invariant polynomials (PIPs), neural networks, and Gaussian approximation potentials (GAPs) in representing water two-body and three-body interaction energies, denoting the resulting potentials PIP-MB-pol, Behler-Parrinello neural network-MB-pol, and GAP-MB-pol, respectively. Our analysis shows that all three analytical representations exhibit similar levels of accuracy in reproducing both two-body and three-body reference data as well as interaction energies of small water clusters obtained from calculations carried out at the coupled cluster level of theory, the current gold standard for chemical accuracy. These results demonstrate the synergy between interatomic potentials formulated in terms of a many-body expansion, such as MB-pol, that are physically sound and transferable, and machine-learning techniques that provide a flexible framework to approximate the short-range interaction energy terms.Keywords
All Related Versions
Funding Information
- National Science Foundation (ACI-1548562, ACI-1642336)
- Deutsche Forschungsgemeinschaft (Be3264/11-2)
- H2020 European Research Council (677013-HBMAP)
This publication has 87 references indexed in Scilit:
- Systematic Improvement of a Classical Molecular Model of WaterThe Journal of Physical Chemistry B, 2013
- Simplifying the representation of complex free-energy landscapes using sketch-mapProceedings of the National Academy of Sciences, 2011
- Permutationally invariant potential energy surfaces in high dimensionalityInternational Reviews in Physical Chemistry, 2009
- CUR matrix decompositions for improved data analysisProceedings of the National Academy of Sciences, 2009
- Determination of Alkali and Halide Monovalent Ion Parameters for Use in Explicitly Solvated Biomolecular SimulationsThe Journal of Physical Chemistry B, 2008
- Water as an Active Constituent in Cell BiologyChemical Reviews, 2007
- Development of transferable interaction models for water. IV. A flexible, all-atom polarizable potential (TTM2-F) based on geometry dependent charges derived from an ab initio monomer dipole moment surfaceThe Journal of Chemical Physics, 2002
- Water–water and water–ion potential functions including terms for many body effectsThe Journal of Chemical Physics, 1985
- Comparison of simple potential functions for simulating liquid waterThe Journal of Chemical Physics, 1983
- Monte Carlo Equation of State of Molecules Interacting with the Lennard-Jones Potential. I. A Supercritical Isotherm at about Twice the Critical TemperatureThe Journal of Chemical Physics, 1957