讲座人简介:

香港中文大学（深圳）数据科学学院助理教授、校长青年学者。此前，方教授于 2020 至 2023 年担任百度量子计算研究所资深研究员，2018 至 2020 年分别在英国剑桥大学和加拿大滑铁卢大学担任博士后研究员。于 2018 年获得悉尼科技大学量子信息方向博士学位，于 2015 年获得武汉大学数学学士学位。方教授于 2022 年入选北京市高层次留学人才回国资助项目，现主持国家自然科学基金重大研究计划（培育）项目和青年项目（C 类）、深圳市面上项目等。截至目前，在 Nature Physics, Physical Review Letters，PRX Quantum，Communications in Mathematical Physics，IEEE Transactions on Information Theory，Mathematical Programming 等物理、数学、计算机以及运筹优化领域的国际顶级学术期刊和会议发表论文二十余篇，在 QIP， TQC，AQIS，ISIT，ICCOPT 等国际顶级学术会议上完成正式报告二十余次。自 2021 年起，方教授完成六十余项国内外技术专利申请，其中五十余项已获得授权。

讲座简介：

We study quantum hypothesis testing with composite correlated hypotheses, a general framework that encompasses scenarios where quantum states are not fully specified and may exhibit correlations beyond the i.i.d. setting. We characterize optimal performance in the error exponent, strong converse exponent, and Chernoff exponent regimes, complementing and refining existing results in the Stein exponent regime, thereby providing an almost complete picture of quantum hypothesis testing in this realistic yet challenging setting. Our results hold under minimal and broadly applicable assumptions — convexity, compactness, and stability under tensor product—conditions satisfied in most practical applications. The generality of these results makes them readily applicable across a wide range of tasks in quantum information theory.