Accelerating equilibrium spin-glass simulations using quantum annealers via generative deep learning

TL;DR

This paper demonstrates how quantum annealers combined with generative neural networks can significantly accelerate equilibrium simulations of spin-glass models, reducing equilibration times compared to traditional methods.

Contribution

The authors introduce a hybrid approach that leverages quantum annealer data and deep learning to improve Markov chain Monte Carlo simulations of spin glasses.

Findings

Hybrid quantum-classical schemes outperform single spin-flip algorithms.

The method achieves shorter equilibration times than traditional Monte Carlo methods.

It is competitive with parallel tempering in reducing correlation times.

Abstract

Adiabatic quantum computers, such as the quantum annealers commercialized by D-Wave Systems Inc., are routinely used to tackle combinatorial optimization problems. In this article, we show how to exploit them to accelerate equilibrium Markov chain Monte Carlo simulations of computationally challenging spin-glass models at low but finite temperatures. This is achieved by training generative neural networks on data produced by a D-Wave quantum annealer, and then using them to generate smart proposals for the Metropolis-Hastings algorithm. In particular, we explore hybrid schemes by combining single spin-flip and neural proposals, as well as D-Wave and classical Monte Carlo training data. The hybrid algorithm outperforms the single spin-flip Metropolis-Hastings algorithm. It is competitive with parallel tempering in terms of correlation times, with the significant benefit of a much shorter equilibration time.