Fermionic neural-network states for ab-initio electronic structure

TL;DR

This paper introduces a neural-network approach for modeling interacting fermionic systems in electronic structure calculations, achieving high accuracy and surpassing traditional methods.

Contribution

It extends neural-network quantum states to fermionic problems by mapping fermionic degrees of freedom to spins, enabling accurate ab-initio electronic structure calculations.

Findings

Achieved near-complete correlation energy recovery for test molecules.

Systematically outperformed coupled cluster and Jastrow wave functions.

Reached chemical accuracy or better in benchmark molecules.

Abstract

Neural-network quantum states have been successfully used to study a variety of lattice and continuous-space problems. Despite a great deal of general methodological developments, representing fermionic matter is however still early research activity. Here we present an extension of neural-network quantum states to model interacting fermionic problems. Borrowing techniques from quantum simulation, we directly map fermionic degrees of freedom to spin ones, and then use neural-network quantum states to perform electronic structure calculations. For several diatomic molecules in a minimal basis set, we benchmark our approach against widely used coupled cluster methods, as well as many-body variational states. On the test molecules, we recover almost the entirety of the correlation energy. We systematically improve upon coupled cluster methods and Jastrow wave functions, reaching levels of chemical accuracy or better. Finally, we discuss routes for future developments and improvements of the methods presented.