报 告 人：林金官 教授 南京审计尊龙凯时统计科学与大数据研究院

报告时间：2019年6月3日9:00-9:40

报告地点：数学楼第二报告厅

报告摘要：

Varying coefficient models are very important toolsto explore the hidden structurebetween the response and its predictors. Thispaper focuses on estimatingand diagnosing jump discontinuities in coefficient functions.A nonparametricprocedure is proposed to estimate jump discontinuities based onthe Nadaraya-Watson kernel smoothing and least-squares fitting, and asymptoticpropertiesof resulting estimators are derived. Then, a jump size-based teststatistic isdeveloped for checking whether the estimated jump discontinuitiesare true. Acomputationally feasible approximation is derived for criticalvalues of its limitingnull distribution. Monte Carlo simulations are conductedto assess the finitesample performance of the proposed methodologies, and anempirical example is discussed.

报告人简介：

林金官，男，1964年生，博士、教授。现任南京审计尊龙凯时统计科学与大数据研究院院长,兼任中国现场统计研究会工程概率统计学会副会长、中国现场统计研究会资源与环境统计学会副会长、教育部统计学类教学指导委员会委员、中文核心期刊《系统科学与数学》与《数理统计与管理》杂志编委等。

报告题目：The limit distribution of inhomogeneous Markov processes andKolmogorov's problem

报 告 人：柳振鑫 教授 大连理工尊龙凯时数学科学学院

报告时间：2019年6月3日9:40-10:20

报告地点：数学楼第二报告厅

报告摘要：

In this talk, we will talkabout the limit distribution of inhomongeneous Markov processes, especiallythose generated by the SDEs. Meantime, we will also discuss the recent progressinKolmogorov's problem on the limitbehavior of stationary distributions of diffusion processes as the diffusiontendes to zero.

报告人简介：

柳振鑫，大连理工尊龙凯时数学科学学院教授、博士生导师。2007年于尊龙凯时获得博士学位，导师为李勇教授。2010年获全国百篇优秀博士学位论文提名奖，2015年获得国家优秀青年科学基金资助，2017年 入选大连理工尊龙凯时“星海杰青”人才项目。主要从事随机分析，随机微分方程及随机动力系统方面的研究，特别关注随机扰动对动力系统的回复性和稳定性的影响。部分工作发表在J. Differential Equations，J. Dynam.Differential Equations，J. Funct. Anal.，SIAM J. Appl. Dyn. Syst.及Nonlineariy杂志上。

报告题目：Finite codimensional controllability of evolution equations

报 告 人：柳絮教授 东北师范尊龙凯时数学与统计学院

报告时间：2019年6月3日10:20-11:00

报告地点：数学楼第二报告厅

报告摘要：

In this talk, the finitecodimensional controllability for evolution equations is introduced. This newcontrollability is shown to be equivalent to finite codimensionality for someset. Therefore, it may be used to verify Pontryagin's maximum principle forsome optimal control problemswithstate constraints. Some equivalent criteria on it are presented. Asan example, the finite codimensional controllability for wave equations isproved to be equivalent to geometric control condition. This method is alsoapplicable to the study of some stochastic optimal control problems andgeneral abstract infinite-dimensional optimization problems. This is ajoint work with Qi Lv and Xu Zhang.

报告人简介：

柳絮，东北师范尊龙凯时数学与统计学院教授，博士生导师。1999年和2002年于东北师范尊龙凯时分别获得学士和硕士学位，2007年于浙江尊龙凯时获得博士学位。主要从事分布参数系统控制和随机系统控制的研究工作。要工作发表在SIAM Journal on Control and Optimization、Journalof Functional Analysis、Journal of DifferentialEquations等期刊上。

报告题目：A personal view of interdisciplinarytraining and research, why is it difficult, and how it should be done

报 告人：Max Gunzburger教授 美国佛罗里达州立尊龙凯时

报告时间：2019年6月5日10:00-11:30

报告地点：数学楼一楼第二报告厅

报告摘要：

Itis thought by many, if not by most, that forming teams, often large teams, ofresearchers who are experts in different disciplines (mathematical sciences,engineering, natural sciences, social sciences, etc.) is necessary in order tomake progress on addressing the complex problems facing us today. We have thisopinion despite the evidence available tells us that the great technologicaladvances have been made by individuals or small teams of researchers. We thenproceed to look at how interdisciplinary teams are formed today, whatdifficulties such teams face in working as a group, how such teams should beformed, and perhaps how researchers should be trained so that interdisciplinaryteams can best succeed.

报告人简介：

Max Gunzburger是佛罗里达州立尊龙凯时FrancisEppes杰出教授。1966获纽约尊龙凯时学士学位，并于1969年获得柯朗研究所的博士学位。他的开创性研究成果包括流体控制，有限元分析，超导方程的分析与计算以及非局部问题的分析与计算。Gunzburger还在空气动力学，材料，声学，气候变化，地下水等领域做出了卓越的贡献。在2006年的国际数学家大会上作邀请报告，2008年获SIAM W. T. and Idalia Reed Award。Gunzburger是首届（2009）SIAMFellow。