CUED Publications database

Hierarchical machine learning of potential energy surfaces.

Dral, PO and Owens, A and Dral, A and Csányi, G (2020) Hierarchical machine learning of potential energy surfaces. J Chem Phys, 152. p. 204110.

Full text not available from this repository.


We present hierarchical machine learning (hML) of highly accurate potential energy surfaces (PESs). Our scheme is based on adding predictions of multiple Δ-machine learning models trained on energies and energy corrections calculated with a hierarchy of quantum chemical methods. Our (semi-)automatic procedure determines the optimal training set size and composition of each constituent machine learning model, simultaneously minimizing the computational effort necessary to achieve the required accuracy of the hML PES. Machine learning models are built using kernel ridge regression, and training points are selected with structure-based sampling. As an illustrative example, hML is applied to a high-level ab initio CH3Cl PES and is shown to significantly reduce the computational cost of generating the PES by a factor of 100 while retaining similar levels of accuracy (errors of ∼1 cm-1).

Item Type: Article
Divisions: Div C > Applied Mechanics
Depositing User: Cron Job
Date Deposited: 05 Jun 2020 20:23
Last Modified: 04 Mar 2021 04:10
DOI: 10.1063/5.0006498