Efficient non-parametric fitting of potential energy surfaces for polyatomic molecules with Gaussian processes

Abstract: We explore the performance of a statistical learning technique based on Gaussian Process (GP) regression as an efficient non-parametric method for constructing multi-dimensional potential energy surfaces (PES) for polyatomic molecules. Using an example of the molecule N$_4$, we show that a realistic GP model of the six-dimensional PES can be constructed with only 240 potential energy points. We construct a series of the GP models and illustrate the convergence of the accuracy of the resulting surfaces as a function of the number of ${\it ab \ initio}$ points. We show that the GP model based on $\sim 1500$ potential energy points achieves the same level of accuracy as the conventional regression fits based on 16,421 points. The GP model of the PES requires no fitting of ${\it ab \ initio}$ data with analytical functions and can be readily extended to surfaces of higher dimensions.