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

TL;DR

This paper introduces a deep learning framework that predicts electronic properties of molecules and materials by learning many-body Green's functions, enabling efficient and transferable quantum chemistry computations.

Contribution

The authors develop a graph neural network that learns the self-energy in many-electron systems, unifying ground and excited state predictions with physical insights, and demonstrating high data efficiency and transferability.

Findings

Achieves competitive accuracy in predicting electronic excitations.

Demonstrates high transferability across molecules and nanomaterials.

Shows effectiveness in bond-breaking and correlation effects.

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.