【主题】Distributed Learning for High Dimensional Semi-parametric Models
【报告人】吕绍高, 副教授
南京审计大学
【时间】 2019年10月21日（星期一）15:30-16:30
【地点】 上海财经大学统计与管理学院大楼1208会议室
【摘要】We consider a two-fold regularized learning procedure for estimating a partially linear regression model in high dimensions. The proposed method is based on the classical divide and conquer strategy for handing big data and each sub-method defined on each subsample consists of a debiased estimation of the double-regularized least squares approach. With the proposed method, we theoretically prove that our global estimators enjoy total learning rates for different estimated components in our model given an appropriate partition on the total data. Specially, our derived results rely on the underlying interactions of different-type estimators with response to multiple-level structures, as well as the ambient dimension of features. Finally, several simulated experiments are implemented to indicate comparable empirical performance of our debiased technique under the distributed setting.
【嘉宾简介】吕绍高，南京审计大学副教授。2011年获得中国科大-香港城市大学联合培养博士，2011年-2018年在西南财经大学工作。主要研究方向是统计机器学习，当前研究兴趣包括分布式学习、随机算法的统计推断以及深度学习的理论分析等。迄今为止在SCI检索的杂志上发表论文20多篇，包括知名期刊 《Annals of Statistics》,《Journal of Machine Learning Research》与《Journal of Econometrics》。主持国家自然科学基金项目3项。长期担任人工智能顶级会议“NeurIPS”、“ICML”、“AAAI”以及“AIStat”程序委员或审稿人。
【主持人】贺莘
