Equivariant neural network for Green's functions of molecules and materials
arXiv:2312.14680 · doi:10.1103/PhysRevB.109.075112
Abstract
The many-body Green's function provides access to electronic properties beyond density functional theory level in ab inito calculations. In this manuscript, we propose a deep learning framework for predicting the finite-temperature Green's function in atomic orbital space, aiming to achieve a balance between accuracy and efficiency. By predicting the self-energy matrices in Lehmann representation using an equivariant message passing neural network, our method respects its analytical property and the equivariance. The Green's function is obtained from the predicted self-energy through Dyson equation with target total number of electrons. We present proof-of-concept benchmark results for both molecules and simple periodic systems, showing that our method is able to provide accurate estimate of physical observables such as energy and density of states based on the predicted Green's function.
References in corpus (36)
- Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
- E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
- Machine Learning of Accurate Energy-Conserving Molecular Force Fields
- Recent developments in the PySCF program package
- Deep-Learning Density Functional Theory Hamiltonian for Efficient ab initio Electronic-Structure Calculation
- Transferability in Machine Learning for Electronic Structure via the Molecular Orbital Basis
- Compressing Green's function using intermediate representation between imaginary-time and real-frequency domains
- Machine learning for many-body physics: The case of the Anderson impurity model
- Dynamical mean-field theory from a quantum chemical perspective
- General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian
- The description of strong correlation within self-consistent Green's function second-order perturbation theory
- Nevanlinna Analytical Continuation
- Sparse sampling approach to efficient ab initio calculations at finite temperature
- A Universal Density Matrix Functional from Molecular Orbital-Based Machine Learning: Transferability across Organic Molecules
- Towards ab initio self-energy embedding theory in quantum chemistry
- Second-order Green's function perturbation theory for periodic systems
- Transferable E(3) equivariant parameterization for Hamiltonian of molecules and solids
- Self-energy embedding theory (SEET) for periodic systems
- Full-Frequency GW without Frequency
- Efficient Formulation of Ab Initio Quantum Embedding in Periodic Systems: Dynamical Mean-Field Theory
- Analytical Continuation of Matrix-Valued Functions: Carathéodory Formalism
- Discrete Lehmann representation of imaginary time Green's functions
- Exploring connections between statistical mechanics and Green's functions for realistic systems. Temperature dependent electronic entropy and internal energy from a self-consistent second-order Green's function
- Fully Self-Consistent Finite-Temperature in Gaussian Bloch Orbitals for Solids
- Chebyshev polynomial representation of imaginary time response functions
- irbasis: Open-source database and software for intermediate-representation basis functions of imaginary-time Green's function
- Robust analytic continuation of Green's functions via projection, pole estimation, and semidefinite relaxation
- Fractional charge and spin errors in self-consistent Green's function theory
- Performance analysis of a physically constructed orthogonal representation of imaginary-time Green's function
- Legendre-spectral Dyson equation solver with super-exponential convergence
- libdlr: Efficient imaginary time calculations using the discrete Lehmann representation
- Effect of propagator renormalization on the band gap of insulating solids
- Efficient excitations and spectra within a perturbative renormalization approach
- Predicting impurity spectral functions using machine learning
- Excitations and spectra from equilibrium real-time Green's functions
- Lehmann representation of the nonequilibrium self-energy