3-regular three-XORSAT planted solutions benchmark of classical and quantum heuristic optimizers

Abstract With current semiconductor technology reaching its physical limits, special-purpose hardware has emerged as an option to tackle specific computing-intensive challenges. Optimization in the form of solving quadratic unconstrained binary optimization problems, or equivalently Ising spin glasses, been focus several new dedicated platforms. These platforms come many different flavors, from highly-efficient implementations on digital-logic established algorithms proposals analog implementing algorithms. In this work, we use a mapping class linear equations whose solutions can be found efficiently, hard constraint satisfaction problem (three-regular three-XORSAT, glass) with ‘golf-course’ shaped energy landscape, benchmark these approaches. We perform scaling and prefactor analysis performance Fujitsu’s digital annealer unit (DAU), D-Wave advantage quantum annealer, virtual MemComputing machine, Toshiba’s simulated bifurcation machine (SBM), SATonGPU algorithm Bernaschi et al , our implementation parallel tempering. identify DAU currently having smallest exponent for benchmark, small addition by far thanks massive parallelism. Our work provides objective assessment snapshot promise limitations relative particular problems.