论文标题
用于估计量子熵的量子算法
Quantum algorithms for estimating quantum entropies
论文作者
论文摘要
von Neumann和Quantumrényi熵表征了量子系统的基本特性,并导致许多领域的理论和实际应用。文献中已经建立了使用准备输入状态纯化的量子查询模型来估计量子熵的量子算法。 {但是,构建这种模型几乎与状态层析成像一样难。}在本文中,我们提出了量子算法来估算$ n $ qubit量子状态$ρ$的von Neumann和量子$α$-Rényi熵,使用输入状态的独立副本。我们还展示了如何使用原始的单个/两级门有效地构建{量子熵估计}的量子电路。我们证明,所需副本的数量以$ 1/ε$和$ 1/λ$为单位,其中$ε$表示添加精度,$λ$表示所有非零特征值的下限。值得注意的是,我们的方法在实用性方面优于以前的方法,因为它不需要任何量子查询词句,这通常是先前方法所必需的。此外,我们进行实验,以显示算法对单量状态的功效并研究噪声鲁棒性。我们还讨论了对某些实践关注的量子状态的应用以及一些有意义的任务,例如量子吉布斯状态准备和纠缠估算。
The von Neumann and quantum Rényi entropies characterize fundamental properties of quantum systems and lead to theoretical and practical applications in many fields. Quantum algorithms for estimating quantum entropies, using a quantum query model that prepares the purification of the input state, have been established in the literature. {However, constructing such a model is almost as hard as state tomography.} In this paper, we propose quantum algorithms to estimate the von Neumann and quantum $α$-Rényi entropies of an $n$-qubit quantum state $ρ$ using independent copies of the input state. We also show how to efficiently construct the quantum circuits for {quantum entropy estimation} using primitive single/two-qubit gates. We prove that the number of required copies scales polynomially in $1/ε$ and $1/Λ$, where $ε$ denotes the additive precision and $Λ$ denotes the lower bound on all non-zero eigenvalues. Notably, our method outperforms previous methods in the aspect of practicality since it does not require any quantum query oracles, which are usually necessary for previous methods. Furthermore, we conduct experiments to show the efficacy of our algorithms to single-qubit states and study the noise robustness. We also discuss the applications to some quantum states of practical interest as well as some meaningful tasks such as quantum Gibbs state preparation and entanglement estimation.