Network-Initialized Monte Carlo Based on Generative Neural Networks

TL;DR

This paper introduces a neural network-based Monte Carlo method that produces uncorrelated samples, reducing thermalization time and accelerating simulations across various physical models, including challenging many-electron systems.

Contribution

The authors propose a novel neural network initialization scheme for Monte Carlo that eliminates autocorrelation and speeds up simulations, especially in complex quantum systems.

Findings

Generates uncorrelated Monte Carlo configurations without autocorrelation.

Reduces thermalization time in Monte Carlo simulations.

Accelerates simulations of many-electron quantum systems.

Abstract

We design generative neural networks that generate Monte Carlo configurations with complete absence of autocorrelation from which only short Markov chains are needed before making measurements for physical observables, irrespective of the system locating at the classical critical point, fermionic Mott insulator, Dirac semimetal, or quantum critical point. We further propose a network-initialized Monte Carlo scheme based on such neural networks, which provides independent samplings and can accelerate the Monte Carlo simulations by significantly reducing the thermalization process. We demonstrate the performance of our approach on the two-dimensional Ising and fermion Hubbard models, and expect it can systematically speed up the Monte Carlo simulations especially for the very challenging many-electron problems.