Improved accuracy and transferability of molecular-orbital-based machine learning: Organics, transition-metal complexes, non-covalent interactions, and transition states

TL;DR

This paper enhances molecular-orbital-based machine learning (MOB-ML) for predicting molecular energies, emphasizing physical constraints, and demonstrates its high accuracy and transferability across diverse chemical datasets, including transition states and non-covalent interactions.

Contribution

The study introduces physically constrained MOB-ML models that require minimal training data and achieve high accuracy across various chemical systems and extrapolation tasks.

Findings

MOB-ML achieves sub-kcal/mol accuracy with only 1% training data.

The method maintains accuracy when transferred to larger molecules and different datasets.

Active learning with Gaussian process variance effectively extends MOB-ML to new chemical regions.

Abstract

Molecular-orbital-based machine learning (MOB-ML) provides a general framework for the prediction of accurate correlation energies at the cost of obtaining molecular orbitals. We demonstrate the importance of preserving physical constraints, including invariance conditions and size consistency, when generating the input for the machine learning model. Numerical improvements are demonstrated for different data sets covering total and relative energies for thermally accessible organic and transition-metal containing molecules, non-covalent interactions, and transition-state energies. MOB-ML requires training data from only 1% of the QM7b-T data set (i.e., only 70 organic molecules with seven and fewer heavy atoms) to predict the total energy of the remaining 99% of this data set with sub-kcal/mol accuracy. This MOB-ML model is significantly more accurate than other methods when transferred to a data set comprised of thirteen heavy atom molecules, exhibiting no loss of accuracy on a size intensive (i.e., per-electron) basis. It is shown that MOB-ML also works well for extrapolating to transition-state structures, predicting the barrier region for malonaldehyde intramolecular proton-transfer to within 0.35 kcal/mol when only trained on reactant/product-like structures. Finally, the use of the Gaussian process variance enables an active learning strategy for extending MOB-ML model to new regions of chemical space with minimal effort. We demonstrate this active learning strategy by extending a QM7b-T model to describe non-covalent interactions in the protein backbone-backbone interaction data set to an accuracy of 0.28 kcal/mol.