Alchemical and structural distribution based representation for improved QML

TL;DR

This paper presents a new atomic representation for quantum machine learning models that improves property prediction across diverse chemical systems by incorporating elemental and structural information, enabling better extrapolation.

Contribution

It introduces an alchemical and structural distribution-based representation that enhances QML model accuracy and allows for extrapolation to unseen element combinations.

Findings

Favorable learning curves for diverse systems

Enables alchemical extrapolation to new bonds

Improves prediction accuracy for electronic properties

Abstract

We introduce a representation of any atom in any chemical environment for the generation of efficient quantum machine learning (QML) models of common electronic ground-state properties. The representation is based on scaled distribution functions explicitly accounting for elemental and structural degrees of freedom. Resulting QML models afford very favorable learning curves for properties of out-of-sample systems including organic molecules, non-covalently bonded protein side-chains, (H$_2$O)$_{40}$-clusters, as well as diverse crystals. The elemental components help to lower the learning curves, and, through interpolation across the periodic table, even enable "alchemical extrapolation" to covalent bonding between elements not part of training, as evinced for single, double, and triple bonds among main-group elements.