ABSTRACT Quantum annealing (QA) has the potential to solve complex optimization problems, but its broad adoption is hindered by qubit errors, which cause an exponential slowdown in computing the global optimal solution as problem size increases. To address this limitation, we present a novel post‐processing error mitigation for optimization technique, known as SEMO (spin‐error mitigation for optimization), designed to improve QA computational efficiency. Its effectiveness is demonstrated through a correlated 3D image segmentation model for material microstructures, formulated as a quadratic unconstrained binary optimization (QUBO) problem. The technique significantly reduces the computing time required to reach the global optimum solution, demonstrating a time‐to‐solution advantage over the standard D‐Wave Systems Advantage QA, simulated annealing (SA), and other post‐processing techniques. Because it operates at the fundamental QUBO level, this approach is applicable to a wide range of discrete combinatorial optimization problems that can be expressed in QUBO format. As the number of qubits in quantum computers continues to increase, this method will become increasingly impactful.
Yang et al. (Sat,) studied this question.