Universal materials model of deep-learning density functional theory Hamiltonian

TL;DR

This paper develops a universal deep-learning model for density functional theory Hamiltonians, enabling accurate, large-scale materials property predictions across diverse compositions and structures, advancing AI-driven materials discovery.

Contribution

It introduces a universal DeepH model trained on a large database, capable of modeling various materials and structures with high accuracy, and demonstrates fine-tuning for specific applications.

Findings

Achieved high accuracy in predicting material properties across diverse structures.

Developed a universal DeepH model capable of handling multiple elemental compositions.

Showcased fine-tuning to improve specific material predictions.

Abstract

Realizing large materials models has emerged as a critical endeavor for materials research in the new era of artificial intelligence, but how to achieve this fantastic and challenging objective remains elusive. Here, we propose a feasible pathway to address this paramount pursuit by developing universal materials models of deep-learning density functional theory Hamiltonian (DeepH), enabling computational modeling of the complicated structure-property relationship of materials in general. By constructing a large materials database and substantially improving the DeepH method, we obtain a universal materials model of DeepH capable of handling diverse elemental compositions and material structures, achieving remarkable accuracy in predicting material properties. We further showcase a promising application of fine-tuning universal materials models for enhancing specific materials models. This work not only demonstrates the concept of DeepH's universal materials model but also lays the groundwork for developing large materials models, opening up significant opportunities for advancing artificial intelligence-driven materials discovery.