Deep-Learning Density Functional Theory Hamiltonian for Efficient ab initio Electronic-Structure Calculation

Abstract

The marriage of density functional theory (DFT) and deep learning methods has the potential to revolutionize modern computational materials science. Here we develop a deep neural network approach to represent DFT Hamiltonian (DeepH) of crystalline materials, aiming to bypass the computationally demanding self-consistent field iterations of DFT and substantially improve the efficiency of ab initio electronic-structure calculations. A general framework is proposed to deal with the large dimensionality and gauge (or rotation) covariance of DFT Hamiltonian matrix by virtue of locality and is realized by the message passing neural network for deep learning. High accuracy, high efficiency and good transferability of the DeepH method are generally demonstrated for various kinds of material systems and physical properties. The method provides a solution to the accuracy-efficiency dilemma of DFT and opens opportunities to explore large-scale material systems, as evidenced by a promising application to study twisted van der Waals materials.

References in corpus (8)

• Nearsightedness of Electronic Matter
• Fractional Chern insulators in magic-angle twisted bilayer graphene
• SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effects
• Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems
• Deep-Learning Density Functional Theory Hamiltonian for Efficient ab initio Electronic-Structure Calculation
• The crucial role of atomic corrugation on the flat bands and energy gaps of twisted bilayer graphene at the "magic angle" $\theta \sim 1.08^{\circ }$
• Quantum deep field: data-driven wave function, electron density generation, and atomization energy prediction and extrapolation with machine learning
• SE(3)-equivariant prediction of molecular wavefunctions and electronic densities

Cited by in corpus (48)

• DeePMD-kit v2: A software package for Deep Potential models
• Towards self-driving laboratories: The central role of density functional theory in the AI age
• Deep-Learning Density Functional Theory Hamiltonian for Efficient ab initio Electronic-Structure Calculation
• General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian
• Transferable E(3) equivariant parameterization for Hamiltonian of molecules and solids
• AI-driven inverse design of materials: Past, present and future
• Deep-learning electronic-structure calculation of magnetic superstructures
• Deep-learning density functional perturbation theory
• Neural functional theory for inhomogeneous fluids: Fundamentals and applications
• Deep learning tight-binding approach for large-scale electronic simulations at finite temperatures with $ab$ $initio$ accuracy
• Spin-Dependent Graph Neural Network Potential for Magnetic Materials
• Universal materials model of deep-learning density functional theory Hamiltonian
• Universal Machine Learning Kohn-Sham Hamiltonian for Materials
• Overcoming the Barrier of Orbital-Free Density Functional Theory for Molecular Systems Using Deep Learning
• Machine Learning and Data-Driven Methods in Computational Surface and Interface Science
• Neural-network Density Functional Theory Based on Variational Energy Minimization
• Why neural functionals suit statistical mechanics
• Hyper-density functional theory of soft matter
• Neural density functionals: Local learning and pair-correlation matching
• Neural density functional theory of liquid-gas phase coexistence
• Bridging electronic and classical density-functional theory using universal machine-learned functional approximations
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• Unified Deep Learning Framework for Many-Body Quantum Chemistry via Green's Functions
• Exploring the design space of machine-learning models for quantum chemistry with a fully differentiable framework
• Why hyperdensity functionals describe any equilibrium observable
• Picosecond-scale Heterogeneous Melting of Metals at Extreme Non-equilibrium States
• Transferable empirical pseudopotenials from machine learning
• Machine learning based nonlocal kinetic energy density functional for simple metals and alloys
• Evolution of flat bands in MoSe${}_2$/WSe${}_2$ moiré lattices: A study combining machine learning and band unfolding methods
• Adaptive energy reference for machine-learning models of the electronic density of states
• Multi-channel machine learning based nonlocal kinetic energy density functional for semiconductors
• Crystalline Material Discovery in the Era of Artificial Intelligence
• Equivariant neural network for Green's functions of molecules and materials
• KPROJ: A Program for Unfolding Electronic and Phononic Bands
• Superconductor discovery in the emerging paradigm of Materials Informatics
• Discovery of sustainable energy materials via the machine-learned material space
• Integrated workflows and interfaces for data-driven semi-empirical electronic structure calculations
• Equivariant Electronic Hamiltonian Prediction with Many-Body Message Passing
• Review of the tight-binding method applicable to the properties of moiré superlattices
• MoireStudio: A Universal Twisted Electronic Structure Calculation Package
• Many-body mobility edges in 1D and 2D revealed by convolutional neural networks
• Deep-learning atomistic semi-empirical pseudopotential model for nanomaterials
• Predicting Many Crystal Properties via an Adaptive Transformer-based Framework
• Algorithmic differentiation for plane-wave DFT: materials design, error control and learning model parameters
• Fault-tolerant Quantum Chemical Calculations with Improved Machine-Learning Models
• Electronic structures of crystalline and amorphous GeSe and GeSbTe compounds using machine learning empirical pseudopotentials
• Solving deep-learning density functional theory via variational autoencoders
• Acceleration of Atomistic NEGF: Algorithms, Parallelization, and Machine Learning