应数学与统计学院概率统计研究所邀请，华东师范大学统计学院、统计交叉科学研究院张澍一助理教授于2024年11月13日-14日访问兰州大学，期间于11月13日举办专题学术报告。

报告题目：Adaptive semi-supervised inference under quantile risk minimization

时间：11月13日（星期三）下午5:20

地点：理工楼631室

报告摘要：This paper presents a general framework for semi-supervised inference, emphasizing optimality, adaptivity, and debias calibration to enhance robustness. The data are collected from multiple groups, with a small portion having fully labeled observations, while the rest have only unlabeled covariates, which may be either complete or subject to block-wise missingness. The empirical risk function is modeled semiparametrically using an index model with debiasing techniques to maintain model flexibility and dimension reduction. A family of two-step imputation-based semisupervised estimators is proposed, demonstrating improved efficiency compared to their supervised counterparts, particularly under model misspecification (adaptive), leading to powered inference. Specifically, the proposed estimators are proven to achieve semiparametric variance lower bounds (optimal) when the index model is correctly specified. A perturbation resampling procedure is devised for variance estimation. The finite sample performance is evaluated through extensive simulation studies and applications to a financial credit dataset. Although quantile risk minimization is the primary focus of this paper, the proposed methods can be readily adapted for various empirical risk minimization problems involving data with semi-supervised block-wise missing structures.

欢迎广大师生参加！

报告人简介

张澍一，华东师范大学统计学院、统计交叉科学研究院助理教授。北京大学统计学博士，哈佛大学统计系博士后。主要研究方向为大数据分布式学习、多源数据融合、高维检验、统计交叉应用。在《Annals of Statistics》、《Journal of Machine Learning Research》等期刊发表论文十余篇，主持国家自然科学基金青年项目、教育部人文社会科学研究一般项目。入选上海市领军人才（青年海外）、上海市浦江人才计划。担任中国现场统计研究会因果推断分会理事，《Environmetrics》副主编。

甘肃应用数学中心

甘肃省高校应用数学与复杂系统省级重点实验室

兰州大学大数据科学研究中心

兰州大学数学与统计学院

兰州大学萃英学院

二〇二四年十一月十一日