Gaussian Process Models of Potential Energy Surfaces with Boundary Optimisation

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

Abstract

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 long range and the cross-over distance between this model and the Gaussian process is learnt from the training data. Results are presented for different implementations of this procedure, known as boundary optimisation, across the following dimer systems: CONe , HF-Ne, HF-Na + , CO 2-Ne and (CO 2) 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 modelling problems.

Citation

Broad, J., Preston, S., Wheatley, R. J., & Graham, R. S. (2021). Gaussian Process Models of Potential Energy Surfaces with Boundary Optimisation. Journal of Chemical Physics, 155(14), Article 144106. https://doi.org/10.1063/5.0063534

• Journal Article Type: Article
• Acceptance Date: Sep 29, 2021
• Online Publication Date: Oct 13, 2021
• Publication Date: Oct 13, 2021
• Deposit Date: Oct 7, 2021
• Publicly Available Date: Oct 13, 2021
• Journal: Journal of Chemical Physics
• Print ISSN: 0021-9606
• Publisher: American Institute of Physics
• Peer Reviewed: Peer Reviewed
• Volume: 155
• Issue: 14
• Article Number: 144106
• DOI: https://doi.org/10.1063/5.0063534
• Public URL: https://nottingham-repository.worktribe.com/output/6395789
• Publisher URL: https://aip.scitation.org/doi/10.1063/5.0063534
• 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 Broad, J., Preston, S., Wheatley, R. J., & Graham, R. S. (2021). Gaussian Process Models of Potential Energy Surfaces with Boundary Optimisation. Journal of Chemical Physics, 155(14) and may be found at https://aip.scitation.org/doi/10.1063/5.0063534.