Skip to main content

Research Repository

Advanced Search

Gaussian process models of potential energy surfaces with boundary optimization

Broad, Jack; Preston, Simon; Wheatley, Richard J.; Graham, Richard S.

Gaussian process models of potential energy surfaces with boundary optimization Thumbnail


Jack Broad

Professor of Statistics and Applied Mathematics

Profile Image

Associate Professor & Reader in Theoretical Chemistry

Professor of Applied Mathematics


A strategy is outlined to reduce the number of training points required to model intermolecular potentials using Gaussian processes, without reducing accuracy. An asymptotic function is used at a long range, and the crossover distance between this model and the Gaussian process is learnt from the training data. The results are presented for different implementations of this procedure, known as boundary optimization, across the following dimer systems: CO-Ne, HF-Ne, HF-Na+, CO2-Ne, and (CO2)2. The technique reduces the number of training points, at fixed accuracy, by up to ∼49%, compared to our previous work based on a sequential learning technique. The approach is readily transferable to other statistical methods of prediction or modeling problems.

Journal Article Type Article
Acceptance Date Sep 29, 2021
Online Publication Date Oct 13, 2021
Publication Date Oct 14, 2021
Deposit Date Oct 14, 2021
Publicly Available Date Oct 15, 2021
Journal Journal of Chemical Physics
Print ISSN 0021-9606
Electronic ISSN 1089-7690
Publisher AIP Publishing
Peer Reviewed Peer Reviewed
Volume 155
Issue 14
Article Number 144106
Keywords Physical and Theoretical Chemistry; General Physics and Astronomy
Public URL
Publisher URL
Additional Information This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in Jack Broad, Simon Preston, Richard J. Wheatley, and Richard S. Graham
, "Gaussian process models of potential energy surfaces with boundary optimization", J. Chem. Phys. 155, 144106 (2021) and may be found at


You might also like

Downloadable Citations