统计与数据科学系系列学术报告之四百四十七-四百四十八期

时间：2024年8月1日（周四）15:00-16:00

地点：史带楼301室

主持人：复旦大学管理学院统计与数据科学系朱仲义教授

报告人：Professor Feifang Hu

Department of Statistics George Washington University

题目：Statistics and AI: past, current and future

摘要：What is statistics? Different generations may have different answers. Personally, I think: “Statistics is a game among human and nature through data”. This reflects the dynamic interplay of understanding, modeling, and predicting natural phenomena using statistical tools. By leveraging data, humans can decode nature's complexities, make informed decisions, and drive scientific and technological progress. The ongoing advancements in data collection, analysis, and interpretation continue to enhance our ability to play this intricate game with nature. In this talk, I will use several examples (from past, current and possible future) to illustrate the games among human and nature through data. I will also share some stories about Professor Zhang and his personal view of Statistics.

个人简介：胡飞芳，现为乔治·华盛顿大学教授，1994年英属哥伦比亚大学取得博士学位。其研究领域包括：A / B测试：临床试验的适应性设计与推理；生物统计学;Bootstrap方法;设计大数据研究;精准医疗中的统计问题。现为国际统计学顶刊JASA的副主编。他是美国统计学会（ASA），国际数理统计学会（IMS）的Fellow。2004年获得NSF的Career Award。专著2部，已在包括《Biometrika》《JASA》《AoS》等在内的国际统计学顶刊上公开发表学术论文近120篇。

时间：2024年8月1日（周四）16:00-17:00

地点：史带楼301室

主持人：复旦大学管理学院统计与数据科学系朱仲义教授

报告人：Professor Zhang Li-Xin

Zhejiang Gongshang University and Zhejiang University

题目：Why do the randomized play-the-winner rule，the doubly adaptive biased design, and the elephant random walk behave similar?

摘要：The randomized play-the-winner rule (RPW) and doubly adaptive biased coin design are response-adaptive designs (DBCD) proposed for sequentially randomizing patients to treatments in a two-treatment clinical trial so that more patients are assigned to the better treatment as the clinical trial goes on. The elephant random walk (ERW) proposed by Schutz and Trimper (2004) is a non-Markovian discrete-time random walk that has a link to a famous saying that elephants can always remember where they have been. The asymptotic behaviors of RPW rule, DBCD, and ERW have been studied in litterateurs independently, and their asymptotic behaviors are very similar. We present the link between the ERW, the RPW rule, and the DBCD by the recursive stochastic algorithm. With the help of the recursive stochastic algorithm, we obtain the strong invariance principle of multi-dimensional varying-memory ERW with random step sizes. The precise law of the iterated logarithm and the Chung-type law of the iterated logarithm are obtained by applying the strong invariance principle.

个人简介：浙江工商大学特聘教授，浙江大学求是特聘教授。1995年获复旦大学理学博士学位， 1997年晋升为教授，2001年起先后担任浙江大学统计学研究所副所长、常务副所长、所长，浙江大学数学系副主任、数学科学学院副院长。现任浙江大学数据科学研究中心副主任、中国现场统计研究会常务理事、浙江省现场统计研究会理事长。主要从事临床试验自适应随机化设计、概率极限理论、相依数据模型等领域的研究，发表了学术论文170余篇，先后主持国家自然科学基金面上项目5项、杰出青年基金项目1项、重点项目1项、联合基金重点项目1项，于2008年入选教育部“新世纪优秀人才支持计划”，2018年入选浙江省科技创新领军人才，2020年当选Institute of Mathematical Statistics Fellow。