Experimental investigation of performance differences between coherent Ising machines and a quantum annealer

Abstract

Physical annealing systems provide heuristic approaches to solving combinatorial optimization problems. Here, we benchmark two types of annealing machines—a quantum annealer built by D-Wave Systems and measurement-feedback coherent Ising machines (CIMs) based on optical parametric oscillators—on two problem classes, the Sherrington-Kirkpatrick (SK) model and MAX-CUT. The D-Wave quantum annealer outperforms the CIMs on MAX-CUT on cubic graphs. On denser problems, however, we observe an exponential penalty for the quantum annealer [exp(–αDWN2)] relative to CIMs [exp(–αCIMN)] for fixed anneal times, both on the SK model and on 50% edge density MAX-CUT. This leads to a several orders of magnitude time-to-solution difference for instances with over 50 vertices. An optimal–annealing time analysis is also consistent with a substantial projected performance difference. The difference in performance between the sparsely connected D-Wave machine and the fully-connected CIMs provides strong experimental support for efforts to increase the connectivity of quantum annealers.

Leveraging quantum annealing for large MIMO processing in centralized radio access networks

“…Second, there are changes in QA architecture expected in annealers due this year [21] featuring qubits with 2× the degree of Chimera, 2× the number of qubits and with longer range couplings. Based on similar gains in recent results on different problem domains [29], we anticipate this will permit ML problems of size, e.g. 175 × 175 for QPSK and dramatically increase the parallelization opportunity of the chip due to the reduced embedding overhead where each chain now only requires N /12 + 1 qubits.…”

“…By means of extrapolation of improvement trends it is expected that quantum engineering advances in superconducting qubit technology will enable QuAMax to be viable within a decade. Moreover, QuAMax's Ising form (in Section 3.2.2) can be adapted to be run in other emerging physics-based optimization devices based on photonic technologies [29] whose processing times overhead are in principle much faster. Hence, we leave an end-to-end evaluation in a fully centralized RAN architecture, with more advanced hardware, as future work.…”

Not-so-adiabatic quantum computation for the shortest vector problem

“…In a sub-adiabatic regime, however, modelling eigenstate transitions probabilistically could be a natural progression for theoretical analysis of AQC-style algorithms. The development of quantum hardware that can realise this generalised Bose-Hubbard Hamiltonian and assume particle-particle offsite interactions is a target for experimental physicists, and generalising AQC-style algorithms to run on different hardware -such as coherent Ising machines [35,36] -will become increasingly investigated as progress continues towards a post-quantum world. values for P, Q substituted in:…”

Model Predictive Control for Finite Input Systems using the D-Wave Quantum Annealer

“…(5). In addition to the limited number of variables, it is known that the quality of the solution degrades when the original problem has a dense structure 34 . To check the sparseness of our formulated problem, we examine the value of all components of the QUBO matrix [J i j in Eq.…”