Variational quantum algorithm for experimental photonic multiparameter estimation

Abstract

Variational quantum metrology represents a powerful tool for optimizing generic estimation strategies, combining the principles of variational optimization with the techniques of quantum metrology. Such optimization procedures result particularly effective for multiparameter estimation problems, where traditional approaches, requiring prior knowledge of the system behavior, often suffer from limitations due to the curse of dimensionality and computational complexity. To overcome these challenges, we develop a variational approach able to efficiently optimize a multiparameter quantum phase sensor operating in a noisy environment. By exploiting the high reconfigurability of an integrated photonic device, we implement a hybrid quantum-classical feedback loop able to enhance the estimation performances, combining classical optimization techniques with quantum circuit evaluations. The latter allows us to compute the system partial derivatives with respect to the variational parameters by applying the parameter-shift rule, and thus reconstruct experimentally the Fisher information matrix. This in turn is adopted as the cost function of a derivative-free classical learning algorithm run to optimize the measurement settings. Our experimental results reveal significant improvements in terms of estimation accuracy and noise robustness, highlighting the potential of the implementation of variational techniques for practical applications in quantum sensing and more generally for quantum information processing with photonic circuits.