一、报告题目:Stochastic Approximation Based Confidence Regions for Stochastic Variational Inequalities
二、报告人:大连理工大学刘永朝教授
三、报告时间:2022年4月9日星期六下午15:50
四、报告平台:腾讯会议
会议ID:912 762 151
五、摘要:The sample average approximation (SAA) and the stochastic approximation (SA)are two popular schemes for solving the stochastic variational inequalities problem (SVIP). It is of fundamental interest todevelop confidence regions of the true solution to the SVIP when the SA scheme is employed.
In thistalk, we discuss the framework of constructing asymptotic confidence regions for thetrue solution of SVIP with a focus on stochastic dual average method. We firstintroducetheasymptotic normality of the SA solutions both in ergodic sense and non-ergodic sense. Then theonline methods of estimating the covariance matrices in the normal distributions arediscussed.Finally, practical procedures of building the asymptotic confidence regions of solutions to SVIPwith numerical simulations are presented.
六、报告人简介:
刘永朝,博士,教授,博士生导师,大连理工大学数学科学学院副院长。2005年和2008年于大连海事大学数学系获得学士和硕士学位,2011年于大连理工大学数学科学学院获得博士学位,2014年11月至2016年4月在英国南安普顿大学从事博士后研究。主要研究方向为随机优化和分布鲁棒优化,在 Mathematical Programming, SIAM Journal on Optimization, Mathematics of Operations Research,SIAM Journal on Numerical Analysis 等优化顶级期刊发表SCI论文20余篇,主持国家自然科学基金3项。