报告题目：High-Precision Fidelity Estimation with Common Randomized Measurements

姓名：朱黄俊

单位：复旦大学物理学系, 教授

报告地点:数学楼305

报告摘要：

Efficient fidelity estimation of multiqubit quantum states is crucial to quantum information processing. However, existing protocols require order $1/\epsilon^2$ different circuits to estimate the infidelity $\epsilon$ to multiplicative precision, posing a major bottleneck for high-precision fidelity estimation. Here, we prove that Clifford-based common randomized measurement (CRM) shadow estimation achieves a quadratic reduction to $1/\epsilon$ circuits. This follows from tight variance bounds for arbitrary observables under Clifford CRM, controlled by the deviation between the true state and a chosen prior. Under dominant noise models---including depolarizing and Pauli noise---the circuit cost collapses to a constant, independent of both $\epsilon$ and the qubit number, and a single circuit often suffices for intermediate and large systems. We further show that experimentally simpler Clifford measurements outperform $4$-design measurements in many practical scenarios, while both offer exponential advantages over Pauli measurements. In addition, we determine the minimal number of measurement settings required to achieve optimal scaling behavior.

专家简介：

朱黄俊，先后在浙江大学物理系、北京大学物理足球吧 、和新加坡国立大学量子技术中心获学士、硕士、和博士学位。之后在加拿大圆周理论物理研究所和德国科隆大学理论物理研究所从事博士后研究。2018年1月入职复旦大学物理学系，历任研究员、教授。主要研究领域为量子信息基础理论，包括量子测量、量子刻画与验证、纠缠理论、和盲量子计算等。以第一或通讯作者在国际知名期刊发表论文50余篇, 包括PRL 12篇，PRX Quantum 2篇，Nat. Commun. 1篇，和npj Quantum Inf. 4篇。