校庆系列学术报告

报告题目：Robust Graph Change-point Detection for Brain Evolvement Study

报告人：  汪洪浪助理教授  

报告时间：2018年6月25日（星期一）上午10：00—11：00

报告地点：院307报告厅

数学与统计学院

2018.6.22

摘要：This paper studies brainstructural evolvement from resting-state functional magnetic resonance imaging.The brain structure is characterized by a series of Gaussian graphical models,and we propose a robust data-driven method for inferring the structural changesof multiple graphs. The graphs correspond to different subjects, are alignedby, e.g., the ages of the subjects, and need to be estimated from the subjectlevel data. We propose to estimate the structural changes of these graphsthrough a three-step procedure. First, we employ a kernel-smoothing approach toestimate multiple graphs at different ages simultaneously. Secondly, wesummarize graphical information, such as the number of edges, global and localefficiency, for each estimated graph, and align them as a curve. Lastly, wepropose a robust least-absolute-deviation (LAD) type penalization procedurewith the fused Lasso (FL) penalty, named LAD-FL, to infer the change-points inthose graph summary metrics. Our method is theoretically well understood, andresults show that it could effectively capture the brain evolvement pattern.

报告人简介：汪洪浪，美国印第安纳大学-普渡大学、印第安纳波利斯分校（IndianaUniversity-Purdue University Indianapolis，IUPUI）统计系助理教授。研究兴趣：纵向数据和函数型数据的统计分析、高维数据的统计推断与应用、非参数与半参数统计统计分析、经验似然统计推断方法及应用、统计遗传与基因组学。