Transferability in machine-learning for electronic structure via the molecular orbital basis

Abstract

We present a machine learning (ML) method for predicting electronic structure correlation energies using Hartree-Fock input. The total correlation energy is expressed in terms of individual and pair contributions from localized occupied MOs, and Gaussian process regression is used to predict these contributions from a feature set that is based on MO properties, such as Fock, Coulomb, and exchange matrix elements. With the aim of maximizing transferability across chem. systems and compactness of the feature set, we avoid the usual specification of ML features in terms of atom- or geometry-specific information, such atom/element-types, bond-types, or local mol. structure. ML predictions of MP2, CCSD, and CCSD(T) energies are presented for a range of systems, demonstrating that the method maintains accuracy while providing transferability both within and across chem. families; this includes predictions for mols. with atom-types and elements that are not included in the training set. The method holds promise both in its current form and as a proof-of-principle for the use of ML in the design of generalized d.-matrix functionals.