Transferable empirical pseudopotenials from machine learning

Abstract

Machine learning is used to generate empirical pseudopotentials that characterize the local screened interactions in the Kohn-Sham Hamiltonian. Our approach incorporates momentum-range-separated rotation-covariant descriptors to capture crystal symmetries as well as crucial directional information of bonds, thus realizing accurate descriptions of anisotropic solids. Trained empirical potentials are shown to be versatile and transferable such that the calculated energy bands and wave functions without cumbersome self-consistency reproduce conventional ab initio results even for semiconductors with defects, thus fostering faster and faithful data-driven materials researches.

arXiv comment10 pages, 9 figures, 3 tables

References in corpus (27)

• Quantum ESPRESSO: a modular and open-source software project for quantum simulations of materials
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• Machine learning and the physical sciences
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• The PseudoDojo: Training and grading a 85 element optimized norm-conserving pseudopotential table
• DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• DScribe: Library of Descriptors for Machine Learning in Materials Science
• Finding Density Functionals with Machine Learning
• A Density Matrix-based Algorithm for Solving Eigenvalue Problems
• Deep-Learning Density Functional Theory Hamiltonian for Efficient ab initio Electronic-Structure Calculation
• Extended DFT+U+V method with on-site and inter-site electronic interactions
• A learning scheme to predict atomic forces and accelerate materials simulations
• The 2021 Quantum Materials Roadmap
• Optimizing many-body atomic descriptors for enhanced computational performance of machine learning based interatomic potentials
• General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian
• Deep Learning and Density Functional Theory
• DeePKS: a comprehensive data-driven approach towards chemically accurate density functional theory
• Quantum deep field: data-driven wave function, electron density generation, and atomization energy prediction and extrapolation with machine learning
• Parameter-free hybrid functional based on an extended Hubbard model: DFT+U+V
• Efficient First-Principles Approach with a Pseudohybrid Density Functional for Extended Hubbard Interactions
• A Novel Approach to Describe Chemical Environments in High Dimensional Neural Network Potentials
• Accelerated identification of equilibrium structures of multicomponent inorganic crystals using machine learning potentials
• Ab initio study of lattice dynamics of group IV semiconductors using pseudohybrid functionals for extended Hubbard interactions

Cited by in corpus (4)

• Neural-network-supported basis optimizer for the configuration interaction problem in quantum many-body clusters: Feasibility study and numerical proof
• The Enduring Relevance of Semiempirical Quantum Mechanics
• Deep-learning atomistic semi-empirical pseudopotential model for nanomaterials
• Electronic structures of crystalline and amorphous GeSe and GeSbTe compounds using machine learning empirical pseudopotentials