A machine learning method is proposed for representing the elements of diabatic potential energy matrices (PEMs) with high fidelity. This is an extension of the so-called permutation invariant polynomial-neural network (PIP-NN) method for representing adiabatic potential energy surfaces. While for one-dimensional irreducible representations the diagonal elements of a diabatic PEM are invariant under exchange of identical nuclei in a molecular system, the off-diagonal elements require special symmetry consideration, particularly in the presence of a conical intersection. A multiplicative factor is introduced to take into consideration the particular symmetry properties while maintaining the PIP-NN framework. Here, we demonstrate here that the extended PIP-NN approach is accurate in representing diabatic PEMs, as evidenced by small fitting errors and by the reproduction of absorption spectra and product branching ratios in both H2O$$(\tilde{X} / \tilde{B})$$ and NH3$$(\tilde{X} $$/Ã) non-adiabatic photodissociation.
Xie, Changjian, et al. "Permutation invariant polynomial neural network approach to fitting potential energy surfaces. IV. Coupled diabatic potential energy matrices." Journal of Chemical Physics, vol. 149, no. 14, Oct. 2018. https://doi.org/10.1063/1.5054310
Xie, Changjian, Zhu, Xiaolei, Yarkony, David R., & Guo, Hua (2018). Permutation invariant polynomial neural network approach to fitting potential energy surfaces. IV. Coupled diabatic potential energy matrices. Journal of Chemical Physics, 149(14). https://doi.org/10.1063/1.5054310
Xie, Changjian, Zhu, Xiaolei, Yarkony, David R., et al., "Permutation invariant polynomial neural network approach to fitting potential energy surfaces. IV. Coupled diabatic potential energy matrices," Journal of Chemical Physics 149, no. 14 (2018), https://doi.org/10.1063/1.5054310
@article{osti_1612363,
author = {Xie, Changjian and Zhu, Xiaolei and Yarkony, David R. and Guo, Hua},
title = {Permutation invariant polynomial neural network approach to fitting potential energy surfaces. IV. Coupled diabatic potential energy matrices},
annote = {A machine learning method is proposed for representing the elements of diabatic potential energy matrices (PEMs) with high fidelity. This is an extension of the so-called permutation invariant polynomial-neural network (PIP-NN) method for representing adiabatic potential energy surfaces. While for one-dimensional irreducible representations the diagonal elements of a diabatic PEM are invariant under exchange of identical nuclei in a molecular system, the off-diagonal elements require special symmetry consideration, particularly in the presence of a conical intersection. A multiplicative factor is introduced to take into consideration the particular symmetry properties while maintaining the PIP-NN framework. Here, we demonstrate here that the extended PIP-NN approach is accurate in representing diabatic PEMs, as evidenced by small fitting errors and by the reproduction of absorption spectra and product branching ratios in both H2O$(\tilde{X} / \tilde{B})$ and NH3$(\tilde{X} $/Ã) non-adiabatic photodissociation.},
doi = {10.1063/1.5054310},
url = {https://www.osti.gov/biblio/1612363},
journal = {Journal of Chemical Physics},
issn = {ISSN 0021-9606},
number = {14},
volume = {149},
place = {United States},
publisher = {American Institute of Physics (AIP)},
year = {2018},
month = {10}}
Longuet-Higgins, Hugh Christopher; Öpik, U.; Lecorney Pryce, Maurice Henry
Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences, Vol. 244, Issue 1236, p. 1-16https://doi.org/10.1098/rspa.1958.0022