Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems

TL;DR

This paper introduces a formalism and a tensor kernel for machine learning tensorial properties of molecules, respecting rotational symmetries, demonstrated on electrical response predictions of water oligomers.

Contribution

It develops a novel tensorial kernel formalism that generalizes SOAP for predicting tensorial molecular properties with rotational symmetry considerations.

Findings

Successfully predicts electrical responses of water oligomers.

Demonstrates the method's accuracy across different molecular complexities.

Generalizes scalar property kernels to tensorial properties.

Abstract

Statistical learning methods show great promise in providing an accurate prediction of materials and molecular properties, while minimizing the need for computationally demanding electronic structure calculations. The accuracy and transferability of these models are increased significantly by encoding into the learning procedure the fundamental symmetries of rotational and permutational invariance of scalar properties. However, the prediction of tensorial properties requires that the model respects the appropriate geometric transformations, rather than invariance, when the reference frame is rotated. We introduce a formalism that can be used to perform machine-learning of tensorial properties of arbitrary rank for general molecular geometries. To demonstrate it, we derive a tensor kernel adapted to rotational symmetry, which is the natural generalization of the smooth overlap of atomic positions (SOAP) kernel commonly used for the prediction of scalar properties at the atomic scale. The performance and generality of the approach is demonstrated by learning the instantaneous electrical response of water oligomers of increasing complexity, from the isolated molecule to the condensed phase.