arXiv:2610.29146
Evaluation of the quantum Fisher information for large systems via an ancilla
Qi Liu, Xin-Yu Hao, Jing Liu
Abstract
Quantum Fisher information (QFI) is a fundamental quantity in quantum information and quantum mechanics, and evaluating it in various physical systems has become an important research topic. However, its evaluation in large systems still faces many challenges. Hereby we propose an ancilla-based protocol for evaluating the QFI of large systems and establish the conditions under which this protocol can ``perfectly'' evaluate the QFI in theory. As a demonstration, the protocol is applied to quantum circuits to evaluate the QFI of the Greenberger-Horne-Zeilinger state.
AI Summary
The paper asks when the quantum Fisher information (QFI) of a large system can be preserved in a low-dimensional ancilla, allowing QFI evaluation through ancilla measurements. It considers parameter-independent controlled unitaries $U=\sum_i |i\rangle\langle i|\otimes T_i$, which produce an ancilla state $\tilde\rho_a=\sum_i p_i(x)T_i\rho_aT_i^\dagger$, where $p_i=\langle i|\rho(x)|i\rangle$. The analysis uses $F(\tilde\rho_a)\leq I(\{p_i\})\leq F(\rho)$. A proof and saturation condition are given for the extended-convexity inequality $F(\sum_i p_i\rho_i)\leq I(\{p_i\})+\sum_i p_iF(\rho_i)$, using a classical flag register and QFI monotonicity. For an optimal system measurement and a pure ancilla, exact preservation requires ancilla states associated with different scores $\partial_x\ln p_i$ to be orthogonal. If the number $m$ of distinct nonzero scores exceeds the ancilla dimension $n$, exact preservation is impossible. A qubit ancilla suffices theoretically for pure-state families using an SLD eigenbasis. The paper also derives a general mixed-state sub-QFI expression and establishes its equality with QFI for qubits. The circuit example transfers the phase QFI $N_s^2$ of an $N_s$-qubit GHZ state to one ancilla through parity encoding, using $N_s$ controlled gates. The stated practical limitation is linear circuit depth and accumulated gate noise, which may impair performance for large systems. The demonstration is theoretical; optimizing information retention when exact transfer is impossible remains open.
Central claim may not hold: The perfect-transfer construction counts only nonzero score values, although positive-probability zero-score outcomes also require an orthogonal ancilla sector, invalidating the claimed sufficiency of m≤n. Theorem 1 incorrectly identifies a support-projected matrix-element condition as equivalent to its full eigenvector condition; the projected condition can hold even when the QFI bound is strictly unsaturated. The GHZ comparison overlooks direct estimation of the binary parity statistic, so it does not establish the claimed unavoidable growth in statistical measurement complexity, although direct readout still involves all system qubits.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 10 pages, 3 figures