A Nested Molecule-Independent Neural Network Approach for High-Quality Potential Fits
Authors:
- 16 December 2005
- journal article
- research article
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 110 (16) , 5295-5304
Abstract
It is shown that neural networks (NNs) are efficient and effective tools for fitting potential energy surfaces. For H2O, a simple NN approach works very well. To fit surfaces for HOOH and H2CO, we develop a nested neural network technique in which we first fit an approximate NN potential and then use another NN to fit the difference of the true potential and the approximate potential. The root-mean-square error (RMSE) of the H2O surface is 1 cm-1. For the 6-D HOOH and H2CO surfaces, the nested approach does almost as well attaining a RMSE of 2 cm-1. The quality of the NN surfaces is verified by calculating vibrational spectra. For all three molecules, most of the low-lying levels are within 1 cm-1 of the exact results. On the basis of these results, we propose that the nested NN approach be considered a method of choice for both simple potentials, for which it is relatively easy to guess a good fitting function, and complicated (e.g., double well) potentials for which it is much harder to deduce an appropriate fitting function. The number of fitting parameters is only moderately larger for the 6-D than for the 3-D potentials, and for all three molecules, decreasing the desired RMSE increases only slightly the number of required fitting parameters (nodes). NN methods, and in particular the nested approach we propose, should be good universal potential fitting tools.Keywords
This publication has 127 references indexed in Scilit:
- A model of evolution and learningNeural Networks, 2005
- Interpolating moving least-squares methods for fitting potential energy surfaces: Analysis of an application to a six-dimensional systemThe Journal of Chemical Physics, 2004
- Numerical solution of the Schrödinger equation by neural network and genetic algorithmComputer Physics Communications, 2001
- Vibrational eigenstates of four-atom molecules: a parallel strategy employing the implicitly restarted Lanczos methodComputer Physics Communications, 1998
- Artificial neural network methods in quantum mechanicsComputer Physics Communications, 1997
- Neural networks as tools to solve problems in physics and chemistryComputer Physics Communications, 1994
- Ab initio calculation of a global potential, vibrational energies, and wave functions for HCN/HNC, and a simulation of the ÖX̃ emission spectrumThe Journal of Chemical Physics, 1993
- Optimal filtering algorithms for fast learning in feedforward neural networksNeural Networks, 1992
- On the approximate realization of continuous mappings by neural networksNeural Networks, 1989
- A b i n i t i o calculations of electronic and vibrational energies of HCO and HOCThe Journal of Chemical Physics, 1986