Deep Charge: A Deep Learning Model of Electron Density from One-Shot Density Functional Theory Calculation

TL;DR

This paper introduces a deep learning model that accurately predicts electron charge density from one-shot density functional theory calculations, preserving physical symmetries and applicable across diverse material structures.

Contribution

The novel deep learning approach effectively captures atomic environments and scales well for large systems, improving charge density predictions from minimal DFT data.

Findings

Accurate charge density predictions across various structures

Model preserves physical symmetries naturally

Efficient analysis of large-scale condensed matter systems

Abstract

Electron charge density is a fundamental physical quantity, determining various properties of matter. In this study, we have proposed a deep-learning model for accurate charge density prediction. Our model naturally preserves physical symmetries and can be effectively trained from one-shot density functional theory calculation toward high accuracy. It captures detailed atomic environment information, ensuring accurate predictions of charge density across bulk, surface, molecules, and amorphous structures. This implementation exhibits excellent scalability and provides efficient analyses of material properties in large-scale condensed matter systems.