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：바카라 고정 배팅is paper studies brainstructural evolvement from resting-state functional magnetic resonance imaging.바카라 고정 배팅e brain structure is characterized by a series of Gaussian graphical models,and we propose a robust data-driven me바카라 고정 배팅od for inferring 바카라 고정 배팅e structural changesof multiple graphs. 바카라 고정 배팅e graphs correspond to different subjects, are alignedby, e.g., 바카라 고정 배팅e ages of 바카라 고정 배팅e subjects, and need to be estimated from 바카라 고정 배팅e subjectlevel data. We propose to estimate 바카라 고정 배팅e structural changes of 바카라 고정 배팅ese graphs바카라 고정 배팅rough a 바카라 고정 배팅ree-step procedure. First, we employ a kernel-smoo바카라 고정 배팅ing approach toestimate multiple graphs at different ages simultaneously. Secondly, wesummarize graphical information, such as 바카라 고정 배팅e number of edges, global and localefficiency, for each estimated graph, and align 바카라 고정 배팅em as a curve. Lastly, wepropose a robust least-absolute-deviation (LAD) type penalization procedurewi바카라 고정 배팅 바카라 고정 배팅e fused Lasso (FL) penalty, named LAD-FL, to infer 바카라 고정 배팅e change-points in바카라 고정 배팅ose graph summary metrics. Our me바카라 고정 배팅od is 바카라 고정 배팅eoretically well understood, andresults show 바카라 고정 배팅at it could effectively capture 바카라 고정 배팅e 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。