特邀厦门大学张庆昭副教授来校作学术报告

发布单位:数学与统计学院编辑:朱亚宾发布时间:2019-12-25浏览量:14

报告题目:Conditional Score Matching for High-Dimensional Partial Graphical Models

时间:2020年1月3日(周五)下午3:00-4:00

地点:藕舫楼818室

报告人:张庆昭 副教授

主持人:来  鹏 副教授

报告摘要:Network construction has been heavily exploited in multivariate data analysis. In many cases,  connections between a large portion of  variables  are of minimal importance. As such,  partial graphs have played an important role in network construction. Due to the existence of a multiplicative normalization constant, the existing construction approaches may bear high  computational cost.  In this paper, we propose conditional score matching for high-dimensional partial graphical models. This approach is uniquely advantageous  by being not influenced by the  multiplicative normalization constant.  Effective computational algorithm is developed, and it is shown that  computational complexity of the proposed method is less than that of those in the literature. Statistical properties are established, and two extensions are explored to incorporate more information and  accommodate more general distributions.  A wide spectrum of simulations and the analysis of a breast cancer gene expression dataset  demonstrate  competitive performance of the proposed methods.

报告人简介:张庆昭, 厦门大学经济学院统计系和王亚南经济研究院副教授、博士生导师。2013年获得中国科学院数学与系统科学研究院概率论与数理统计博士学位,先后在中国科学院大学和美国耶鲁大学进行博士后研究。主要研究方向为高维数据分析、多源数据融合、函数数据分析、统计学习和数据挖掘等,在JASA、Biometrics、Statistica Sinica等期刊发表论文30余篇。国际统计学会推选会员,主持国家自科面上、青年各1项,教育部基金1项。

数学与统计学院

2019年12月30日