The fitting of potential energy surfaces using neural networks: Application to the study of vibrational levels of H3+

A back-propagation neural network is utilized to fit the potential energy surfaces of the H3+ ion, using the ab initio data points of Dykstra and Swope, and the Meyer, Botschwina, and Burton ab initio data points. We used the standard back-propagation formulation and have also proposed a symmetric formulation to account for the symmetry of the H3+ molecule. To test the quality of the fits we computed the vibrational levels using the correlation function quantum Monte Carlo method. We have compared our results with the available experimental results and with results obtained using other potential energy surfaces. The vibrational levels are in very good agreement with the experiment and the back-propagation fitting is of the same quality of the available potential energy surfaces.

[1]  Martin Fodslette Møller,et al.  A scaled conjugate gradient algorithm for fast supervised learning , 1993, Neural Networks.

[2]  Antonio Laganà,et al.  A new bond-order functional form for triatomic molecules , 1985 .

[3]  D. Fotiadis,et al.  Artificial neural network methods in quantum mechanics , 1997, quant-ph/9705029.

[4]  Eduardo D. Sontag.  Sigmoids Distinguish More Efficiently Than Heavisides , 1989, Neural Computation.

[5]  B. R. Upadhyaya,et al.  Application of neural network computing to the solution for the ground-state eigenenergy of two-dimensional harmonic oscillators , 1991 .

[6]  J. Rychlewski,et al.  Sub-microhartree accuracy potential energy surface for H3 + including adiabatic and relativistic effects. II. Rovibrational analysis for H3 + and D3 + , 1998 .

[7]  Bobby G. Sumpter,et al.  Theory and Applications of Neural Computing in Chemical Science , 1994 .

[8]  Dmitrii E. Makarov,et al.  FITTING POTENTIAL-ENERGY SURFACES : A SEARCH IN THE FUNCTION SPACE BY DIRECTED GENETIC PROGRAMMING , 1998 .

[9]  W. A. Lester,et al.  The calculation of excited states with quantum Monte Carlo. II. Vibrational excited states , 1990 .

[10]  Peter Botschwina,et al.  Ab initio calculation of near‐equilibrium potential and multipole moment surfaces and vibrational frequencies of H+3 and its isotopomers , 1986 .

[11]  C. M. Roach,et al.  Fast curve fitting using neural networks , 1992 .

[12]  I. Shavitt,et al.  Calculation and fitting of potential energy and dipole moment surfaces for the water molecule: Fully ab initio determination of vibrational transition energies and band intensities , 1997 .

[13]  David M. Ceperley,et al.  Fixed-node quantum Monte Carlo for molecules , 1982 .

[14]  Mark N. Gibbs,et al.  Combining ab initio computations, neural networks, and diffusion Monte Carlo: An efficient method to treat weakly bound molecules , 1996 .

[15]  J. J. Soares Neto,et al.  The fitting of potential energy surfaces using neural networks. Application to the study of the photodissociation processes , 1998 .

[16]  Steven D. Brown,et al.  Neural network models of potential energy surfaces , 1995 .

[17]  E. Clementi,et al.  Gaussian functions in Hylleraas‐configuration interaction calculations. V. An accurate abinitio H+3 potential‐energy surface , 1990 .

[18]  Michael A. Collins,et al.  Molecular potential energy surfaces by interpolation in Cartesian coordinates , 1998 .

[19]  G. D. Carney,et al.  A comparison of force fields and calculation methods for vibration intervals of isotopic H+3 molecules , 1986 .

[20]  David J. C. MacKay,et al.  Bayesian Interpolation , 1992, Neural Computation.

[21]  Takeshi Oka,et al.  Observation of the infrared spectrum of H 3 , 1980 .

[22]  P. H. Acioli,et al.  Estimating correlation energy of diatomic molecules and atoms with neural networks , 1997, Journal of Computational Chemistry.

[23]  Richard P. Brent,et al.  Fast training algorithms for multilayer neural nets , 1991, IEEE Trans. Neural Networks.

[24]  G. D. Carney,et al.  H3+: Geometry dependence of electronic properties , 1974 .

[25]  Antônio de Pádua Braga,et al.  Artificial neural network applied for predicting rainbow trajectories in atomic and molecular classical collisions , 1997 .

[26]  J. Michael Finlan,et al.  New alternative to the Dunham potential for diatomic molecules , 1973 .

[27]  W. C. Swope,et al.  The H3+ potential surface , 1979 .

[28]  David Mackay,et al.  Probable networks and plausible predictions - a review of practical Bayesian methods for supervised neural networks , 1995 .