时间：2019年6月20日（周四） 上午9：30-10：30
地点：理科大楼A楼1514（中北校区）
题目：A Parsimonious Model for Personalized Dose Finding
报告人：Zhu Ruoqing 伊利诺伊大学香槟分校助理教授
摘要：
Learning an individualized dose rule (IDR) in personalized medicine is a challenging statistical problem. Existing methods for estimating the optimal IDR often suffer from the curse of dimensionality, especially when the IDR is learned nonparametrically using machine learning approaches. We propose a dimension reduction framework to achieve parsimony and interpretability. The framework postulate that the IDR can be represented by a function that relies only on a few linear combinations of the original covariates. Different from the commonly used outcome weighted learning approach, we directly solve an objective function that nonparametrically approximates the value function. Two methods are proposed, a direct learning approach that yields the IDR as commonly desired in personalized medicine, and a pseudo-direct learning approach that focuses more on learning the dimension reduction space. Under regularity assumptions, we provide the convergence rate for the semiparametric estimators and Fisher consistency of the corresponding value function. In both approaches, we use an orthogonality constrained optimization approach on the Stiefel manifold. Performances of the proposed methods are demonstrated through a warfarin pharmacogenetic dataset.
报告人简介：
Zhu Ruoqing，北卡大学教堂山生物统计系获得理学博士学位。2013年至2015年在耶鲁大学生物统计系进修博士后。2015年起在伊利诺伊大学香槟分校统计系任助理教授。2015至今先后时隶属于Carl R. Woese Institute of Genomic Biology, National Center for Supercomputing Applications (NCSA) 和Carle Illinois College of Medicine (CICOM)。2018 年获得NCSA Research Fellowship。 2017年至今在CICOM 担任课程设置委员会成员。研究兴趣包括随机森林，机器学习，个性化医疗，以及机器学习在生物统计和生物医学中的应用。已在统计学国际顶级期刊JASA, Biometrika上发表多篇文章。