Unified Deep Learning Framework for Many-Body Quantum Chemistry via Green's Functions

Abstract

Quantum many-body methods provide a systematic route to computing electronic properties of molecules and materials, but high computational costs restrict their use in large-scale applications. Due to the complexity in many-electron wavefunctions, machine learning models capable of capturing fundamental many-body physics remain limited. Here, we present a deep learning framework targeting the many-body Green's function, which unifies predictions of electronic properties in ground and excited states, while offering physical insights into many-electron correlation effects. By learning the $GW$ or coupled-cluster self-energy from mean-field features, our graph neural network achieves competitive performance in predicting one- and two-particle excitations and quantities derivable from one-particle density matrix. We demonstrate its high data efficiency and good transferability across chemical species, system sizes, molecular conformations, and correlation strengths in bond breaking, through multiple molecular and nanomaterial benchmarks. This work opens up opportunities for utilizing machine learning to solve many-electron problems.

16 pages, 5 figures

References in corpus (15)

• Continuous-time Monte Carlo methods for quantum impurity models
• The GW compendium: A practical guide to theoretical photoemission spectroscopy
• The Bethe-Salpeter Equation Formalism: From Physics to Chemistry
• Machine learning for many-body physics: The case of the Anderson impurity model
• General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian
• Time-step targeting time-dependent and dynamical density matrix renormalization group algorithms with ab initio Hamiltonians
• Low-scaling $GW$ with benchmark accuracy and application to phosphorene nanosheets
• Graph neural networks for materials science and chemistry
• Gaussian-Based Quasiparticle Self-Consistent $GW$ for Periodic Systems
• Algebraic Diagrammatic Construction Theory for Simulating Charged Excited States and Photoelectron Spectra
• Periodic Coupled-Cluster Green's Function for Photoemission Spectra of Realistic Solids
• Efficient Full-frequency GW Calculations using a Lanczos Method
• Interacting-bath dynamical embedding for capturing non-local electron correlation in solids
• Equivariant neural network for Green's functions of molecules and materials
• Energy-Specific Bethe-Salpeter Equation Implementation for Efficient Optical Spectrum Calculations

Cited by in corpus (6)

• How to choose efficiently the size of the Bethe-Salpeter Equation Hamiltonian for accurate exciton calculations on supercells
• Deep-learning atomistic semi-empirical pseudopotential model for nanomaterials
• Electron correlation in semiconductors and insulators via symbolic regression
• Machine learning of electronic structure and atomistic properties from the external potential
• LibppRPA: An Open-Source Library for Particle-Particle Random Phase Approximation
• Machine Learning Green's Functions of Strongly Correlated Hubbard Models