Deep learning enhanced individual nuclear-spin detection

Abstract The detection of nuclear spins using individual electron has enabled diverse opportunities in quantum sensing and information processing. Proof-of-principle experiments have demonstrated atomic-scale imaging nuclear-spin samples controlled multi-qubit registers. However, to image more complex realize larger-scale processors, computerized methods that efficiently automatically characterize spin systems are required. Here, we a deep learning model for automatic identification the single nitrogen-vacancy (NV) centers diamond as sensor. Based on neural network algorithms, develop noise recovery procedures training sequences highly non-linear spectra. We apply these experimentally demonstrate fast 31 around NV center accurately determine hyperfine parameters. Our can be extended larger applicable wide range electron-nuclear interaction strengths. These results pave way towards efficient characterization large spin-qubit