Physics-Inspired Optimization for Quadratic Unconstrained Problems Using a Digital Annealer

TL;DR

The paper presents the Fujitsu Digital Annealer, a hardware-accelerated solver for quadratic unconstrained binary optimization problems, demonstrating significant speedups over traditional algorithms for dense problems, with potential for larger problem sizes.

Contribution

It introduces a digital annealer with a novel parallel-trial scheme and dynamic escape mechanism, achieving substantial speedups and scalability improvements over existing methods.

Findings

Two orders of magnitude speedup for dense spin-glass problems

No speedup observed for sparse 2D problems, explained theoretically

Improved scaling for fully connected problems of moderate difficulty

Abstract

The Fujitsu Digital Annealer (DA) is designed to solve fully connected quadratic unconstrained binary optimization (QUBO) problems. It is implemented on application-specific CMOS hardware and currently solves problems of up to 1024 variables. The DA's algorithm is currently based on simulated annealing; however, it differs from it in its utilization of an efficient parallel-trial scheme and a dynamic escape mechanism. In addition, the DA exploits the massive parallelization that custom application-specific CMOS hardware allows. We compare the performance of the DA to simulated annealing and parallel tempering with isoenergetic cluster moves on two-dimensional and fully connected spin-glass problems with bimodal and Gaussian couplings. These represent the respective limits of sparse versus dense problems, as well as high-degeneracy versus low-degeneracy problems. Our results show that the DA currently exhibits a time-to-solution speedup of roughly two orders of magnitude for fully connected spin-glass problems with bimodal or Gaussian couplings, over the single-core implementations of simulated annealing and parallel tempering Monte Carlo used in this study. The DA does not appear to exhibit a speedup for sparse two-dimensional spin-glass problems, which we explain on theoretical grounds. We also benchmarked an early implementation of the Parallel Tempering DA. Our results suggest an improved scaling over the other algorithms for fully connected problems of average difficulty with bimodal disorder. The next generation of the DA is expected to be able to solve fully connected problems up to 8192 variables in size. This would enable the study of fundamental physics problems and industrial applications that were previously inaccessible using standard computing hardware or special-purpose quantum annealing machines.