A Grover-search based quantum learning scheme for classification

Abstract The hybrid quantum–classical learning scheme provides a prominent way to achieve quantum advantages on near-term devices. A concrete example toward this goal is the neural network (QNN), which has been developed accomplish various supervised tasks such as classification and regression. However, there are two central issues that remain obscure when QNN exploited tasks. First, classifier can well balance computational cost number of measurements performance unexplored. Second, it unclear whether classifiers be applied solve certain problems outperform their classical counterparts. Here we devise Grover-search based (GBLS) address above issues. Notably, most existing QNN-based seamlessly embedded into proposed scheme. key insight behind our proposal reformulating search problem. Numerical simulations exhibit GBLS comparable with other under noise settings, while required dramatically reduced. We further demonstrate potential advantage over in measure query complexity. Our work guidance develop advanced devices opens up an avenue explore