A Simple and Efficient Equivariant Message Passing Neural Network Model for Non-Local Potential Energy Surface

TL;DR

This paper introduces EquiREANN, a simple and efficient equivariant message passing neural network that effectively models non-local potential energy surfaces with high accuracy and minimal additional computational cost.

Contribution

It presents a novel equivariant message passing framework using atomic orbitals, improving non-local interaction modeling in machine learning potentials.

Findings

Accurately describes non-local potential energy variations.

Achieves high accuracy with little extra computational cost.

Provides a generalized approach for equivariant message passing models.

Abstract

Machine learning potentials have become increasingly successful in atomistic simulations. Many of these potentials are based on an atomistic representation in a local environment, but an efficient description of non-local interactions that exceed a common local environment remains a challenge. Herein, we propose a simple and efficient equivariant model, EquiREANN, to effectively represent non-local potential energy surface. It relies on a physically inspired message passing framework, where the fundamental descriptors are linear combination of atomic orbitals, while both invariant orbital coefficients and the equivariant orbital functions are iteratively updated. We demonstrate that this EquiREANN model is able to describe the subtle potential energy variation due to the non-local structural change with high accuracy and little extra computational cost than an invariant message passing model. Our work offers a generalized approach to create equivariant message passing adaptations of other advanced local many-body descriptors.