Efficient protocol to estimate the Quantum Fisher Information Matrix for Commuting-Block Circuits

Abstract: The Quantum Fisher Information Matrix (QFIM) is a fundamental quantity in various subfields of quantum physics. It plays a crucial role in the study of parameterized quantum states, as it quantifies their sensitivity to variations in its parameters. Recently, the QFIM has been successfully employed to enhance the optimization of variational quantum algorithms. However, its practical applicability is often hindered by the high resource requirements for its estimation. In this work, we introduce a novel protocol for computing the off-block-diagonal elements of the QFIM between different layers in a particular class of variational quantum circuits, known as commuting-block circuits. Our approach significantly reduces the quantum resources required, specifically lowering the number of distinct quantum state preparations from $O(m^2)$ to $O(L^2)$, where $m$ is the total number of parameters and $L$ is the number of layers in the circuit. Consequently, our protocol also minimizes the number of classical measurements and post-processing operations needed to estimate the QFIM, leading to a substantial improvement in computational efficiency.