Neural networks for on-the-fly single-shot state classification

Neural networks have proven to be efficient for a number of practical applications ranging from image recognition identifying phase transitions in quantum physics models. In this paper, we investigate the application neural state classification single-shot measurement. We use dispersive readout superconducting transmon circuit demonstrate an increase assignment fidelity both two and three classifications. More importantly, our method is ready on-the-fly data processing without overhead or need large transfer hard drive. addition, capacity trained against experimental imperfections, such as drift local oscillator heterodyne detection scheme.