Improving density matrix electronic structure method by deep learning

TL;DR

This paper introduces a neural network approach to model the DFT density matrix, enhancing the accuracy and generalizability of electronic structure predictions in materials science.

Contribution

It presents a novel neural network method specifically designed for the density matrix, a key but previously unexplored quantity in deep-learning electronic structure calculations.

Findings

High accuracy in predicting electronic properties

Excellent generalizability across different systems

Capable of reproducing charge densities and other properties

Abstract

The combination of deep learning and ab initio materials calculations is emerging as a trending frontier of materials science research, with deep-learning density functional theory (DFT) electronic structure being particularly promising. In this work, we introduce a neural-network method for modeling the DFT density matrix, a fundamental yet previously unexplored quantity in deep-learning electronic structure. Utilizing an advanced neural network framework that leverages the nearsightedness and equivariance properties of the density matrix, the method demonstrates high accuracy and excellent generalizability in multiple example studies, as well as capability to precisely predict charge density and reproduce other electronic structure properties. Given the pivotal role of the density matrix in DFT as well as other computational methods, the current research introduces a novel approach to the deep-learning study of electronic structure properties, opening up new opportunities for deep-learning enhanced computational materials study.