Neural network Gaussian processes as efficient models of potential energy surfaces for polyatomic molecules
Authors:
Open Access
- 10 November 2023
- journal article
- research article
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 4 (4) , 045027
Abstract
Kernel models of potential energy surfaces (PESs) for polyatomic molecules are often restricted by a specific choice of the kernel function. This can be avoided by optimizing the complexity of the kernel function. For regression problems with very expensive data, the functional form of the model kernels can be optimized in the Gaussian process (GP) setting through compositional function search guided by the Bayesian information criterion. However, the compositional kernel search is computationally demanding and relies on greedy strategies, which may yield sub-optimal kernels. An alternative strategy of increasing complexity of GP kernels treats a GP as a Bayesian neural network (NN) with a variable number of hidden layers, which yields NNGP models. Here, we present a direct comparison of GP models with composite kernels and NNGP models for applications aiming at the construction of global PES for polyatomic molecules. We show that NNGP models of PES can be trained much more efficiently and yield better generalization accuracy without relying on any specific form of the kernel function. We illustrate that NNGP models trained by distributions of energy points at low energies produce accurate predictions of PES at high energies. We also illustrate that NNGP models can extrapolate in the input variable space by building the free energy surface of the Heisenberg model trained in the paramagnetic phase and validated in the ferromagnetic phase. By construction, composite kernels yield more accurate models than kernels with a fixed functional form. Therefore, by illustrating that NNGP models outperform GP models with composite kernels, our work suggests that NNGP models should be a preferred choice of kernel models for PES.Keywords
All Related Versions
Funding Information
- Natural Sciences and Engineering Research Council of Canada
This publication has 53 references indexed in Scilit:
- Neural network‐based approaches for building high dimensional and quantum dynamics‐friendly potential energy surfacesInternational Journal of Quantum Chemistry, 2014
- Calculating vibrational spectra with sum of product basis functions without storing full-dimensional vectors or matricesThe Journal of Chemical Physics, 2014
- A global potential energy surface for the H2 + OH ↔ H2O + H reaction using neural networksThe Journal of Chemical Physics, 2013
- Neural network potential-energy surfaces in chemistry: a tool for large-scale simulationsPhysical Chemistry Chemical Physics, 2011
- Momentum average approximation for models with boson-modulated hopping: Role of closed loops in the dynamical generation of a finite quasiparticle massPhysical Review B, 2010
- Potential Energy Surfaces Fitted by Artificial Neural NetworksThe Journal of Physical Chemistry A, 2010
- Generalized Neural-Network Representation of High-Dimensional Potential-Energy SurfacesPhysical Review Letters, 2007
- A random-sampling high dimensional model representation neural network for building potential energy surfacesThe Journal of Chemical Physics, 2006
- A Nested Molecule-Independent Neural Network Approach for High-Quality Potential FitsThe Journal of Physical Chemistry A, 2005
- Reproducing kernel Hilbert space interpolation methods as a paradigm of high dimensional model representations: Application to multidimensional potential energy surface constructionThe Journal of Chemical Physics, 2003