题 目:Review of Distributed Statistical Inference(分布式统计分析)
报告人:张日权教授 华东师范大学统计学院
时 间:2021.12.1 15:00-16:30
地 点:经管学院 335
Abstract
The rapid emergence of massive datasets invarious fields poses a serious challenge to traditional statistical methods.Meanwhile, it provides opportunities for researchers to develop novelalgorithms. Inspired by the idea of divide-and-conquer, various distributedframeworks for statistical estimation and inference have been proposed. Theywere developed to deal with large-scale statistical optimization problems. Thisreport aims to provide a comprehensive review for related literature. Itincludes parametric models, nonparametric models, and other frequently usedmodels. Their key ideas and theoretical properties are summarized. Thetrade-off between communication cost and estimate precision together with otherconcerns are discussed.
张日权,华东师范大学,统计学院院长
教育部统计与数据科学前沿理论及应用重点实验室主任
《应用概率统计》期刊常务副主编
中国现场统计研究会大数据统计分会 理事长
中国商业统计学会数据科学与商业智能分会副理事长
上海统计学会副会长
主要研究方向:大数据统计,金融统计,非/半参数统计,超/高维数据分析,函数型数据分析、统计机器学习
主持多项国家自然科学基金、教育部博士点基金、国家统计局重点项目等研究项目20多项;发表论文130余篇,SCI收录论文100余篇,由科学出版社出版专著1部,教材1部。
先后获得上海市育才奖,上海市自然科学奖三等奖,上海市教学成果一等奖。