On the Atomic Cluster Expansion: interatomic potentials and beyond

TL;DR

The paper discusses the Atomic Cluster Expansion (ACE), a versatile and systematically improvable method for describing atomic environments, highlighting its significance in developing machine learning-based interatomic potentials.

Contribution

It provides an overview of the ACE framework, emphasizing its potential to advance interatomic potential modeling beyond current approaches.

Findings

ACE offers a universal, invariant descriptor for atomic environments.

ACE enables systematic improvements in interatomic potential accuracy.

The framework has significant potential for impact in machine learning-based materials modeling.

Abstract

The Atomic Cluster Expansion (ACE) [R. Drautz, Phys. Rev. B, 99:014104 (2019)] provides a systematically improvable, universal descriptor for the environment of an atom that is invariant to permutation, translation and rotation. ACE is being used extensively in newly emerging interatomic potentials based on machine learning. This commentary discusses the ACE framework and its potential impact.