Permutation invariant polynomial neural network approach to fitting potential energy surfaces
Abstract
A simple, general, and rigorous scheme for adapting permutation symmetry in molecular systems is proposed and tested for fitting global potential energy surfaces using neural networks (NNs). The symmetry adaptation is realized by using low-order permutation invariant polynomials (PIPs) as inputs for the NNs. This so-called PIP-NN approach is applied to the H + H2 and Cl + H2 systems and the analytical potential energy surfaces for these two systems were accurately reproduced by PIP-NN. The accuracy of the NN potential energy surfaces was confirmed by quantum scattering calculations.
- Publication:
-
Journal of Chemical Physics
- Pub Date:
- August 2013
- DOI:
- 10.1063/1.4817187
- Bibcode:
- 2013JChPh.139e4112J
- Keywords:
-
- atom-molecule collisions;
- chlorine;
- hydrogen neutral atoms;
- hydrogen neutral molecules;
- neural nets;
- polynomials;
- potential energy surfaces;
- quantum chemistry;
- 31.50.-x;
- 34.20.-b;
- Potential energy surfaces;
- Interatomic and intermolecular potentials and forces potential energy surfaces for collisions