Machine learning for real-world quantum-enhanced phase estimation

One of the most immediate practical applications of quantum information processing is performing precise quantum measurements. Important examples include the measurement of time with atomic clocks, spatial displacements with optical interferometry, and super-resolved imaging beyond the diffraction limit. Heisenberg\\\'s uncertainty principle provides a fundamental bound on the amount of information a measurement can extract. Measurement schemes employing adaptive feedback constitute a promising strategy for reaching the Heisenberg limit. However, devising adaptive measurement procedures is complicated and often involves clever guesswork.\\r\\n\\r\\nI present an automated technique, based on machine-learning that replaces guesswork by a logical, fully automatic, programmable routine. I explain our method using the example of interferometric phase estimation, which has applications such as atomic clocks and gravitational wave detection. Our algorithm autonomously learns to perform phase estimation based on experimental trial runs, which can be either simulated or performed using a real world experiment. The algorithm does not require prior knowledge about the experiment and is effective even if the quantum system is a black box. Our new technique is robust against loss and decoherence. Furthermore, our algorithm learns to account for systematic experimental imperfections and random noise, thereby making time-consuming error modelling and extensive calibration dispensable. We show that our method outperforms the best known adaptive scheme for interferometric phase estimation.