统计与数据科学系系列学术报告之三百七十九

时    间：6月21日（周二） 14:00-15:00

地    点：腾讯会议   会议号：285-575-886

主持人：夏寅 教授

题    目：Sparse Recovery With Multiple Data Streams: A Sequential Adaptive Testing Approach

报告人：孙文光教授  浙江大学

个人简介：孙文光教授是浙江大学求是特聘教授和数据科学研究中心主任。回国前是美国南加利福尼亚大学（USC）马绍尔商学院的教授。主要研究方向为大范围多重假设检验，选择性推断，经验贝叶斯方法，迁移学习，机器学习的公平性和统计决策理论。曾获美国国家科学基金会杰出青年教授奖(CAREER AWARD)，USC马绍尔商学院杰出研究奖（2次）和Golden Apple最佳教学奖。曾担任JRSS-B以及Journal of Multivariate Analysis的副主编。

报告摘要：Multistage design has been used in a wide range of scientific fields. By allocating sensing resources adaptively, one can effectively eliminate null locations and localize signals with a smaller study budget. We formulate a decision-theoretic framework for simultaneous multistage adaptive testing and study how to minimize the total number of measurements while meeting pre-specified constraints on both the false positive rate (FPR) and missed discovery rate (MDR). The new procedure, which effectively pools information across individual tests using a simultaneous multistage adaptive ranking and thresholding (SMART) approach, achieves precise error rates control and leads to great savings in total study costs. Numerical studies confirm the effectiveness of SMART for controlling both the FPR and MDR and show that it outperforms existing methods. The SMART procedure is illustrated through the analysis of large-scale A/B tests, high-throughput screening and image analysis. This is the joint work with Weinan Wang at Snap, Inc and Bowen Gang at Fudan University.

统计与数据科学系

2022-6-10