A real neural network state for quantum chemistry

TL;DR

This paper introduces a real-valued neural network inspired by RBM for quantum chemistry, demonstrating comparable accuracy and improved convergence when leveraging Hartree-Fock references.

Contribution

A real-valued neural network model for quantum chemistry that matches RBM accuracy and benefits from Hartree-Fock references for faster convergence and higher precision.

Findings

Comparable precision to RBM on prototypical molecules

Accelerated convergence using Hartree-Fock reference

Enhanced energy accuracy with reference-guided optimization

Abstract

The restricted Boltzmann machine (RBM) has been successfully applied to solve the many-electron Schr$\ddot{\text{o}}$dinger equation. In this work we propose a single-layer fully connected neural network adapted from RBM and apply it to study ab initio quantum chemistry problems. Our contribution is two-fold: 1) our neural network only uses real numbers to represent the real electronic wave function, while we obtain comparable precision to RBM for various prototypical molecules; 2) we show that the knowledge of the Hartree-Fock reference state can be used to systematically accelerate the convergence of the variational Monte Carlo algorithm as well as to increase the precision of the final energy.