报告地点:腾讯会议 171-614-824
邀请人:孙海琳教授
报告摘要: We consider distributionally robust optimization (DRO) when the ambiguity set is given by moments for the distributions. The objective and constraints are given by polynomials in decision variables. The DRO can be equivalently reformulated with moment conic constraints. Under some general assumptions, we prove the DRO is equivalent to a linear optimization problem with moment and psd polynomial cones. The Moment-SOS relaxation method is proposed to solve it. Its asymptotic and finite convergence are shown under certain assumptions.
报告人简介: 聂家旺,加州大学圣地亚哥分校教授,湘潭大学兼职教授。他长期从事最优化理论与计算、张量计算和凸代数几何等领域的研究,做出了一系列突破性的工作,主要的学术成果包括:提出了求解多项式优化全局最优解的精确松弛系列并证明了它的紧性;给出了凸半代数集可被SDP表示的充分性与必要性条件;建立了多项式优化中局部最优性条件与全局最优性条件之间的联系; 给出了多项式优化中的拉格朗日乘子表示定理;提出了生成多项式作为计算张量分解和低秩逼近的新方法。他在《Mathematical Programming》、《Foundations of Computational Mathematics》,SIAM系列杂志等顶级学术期刊上,发表论文60余篇,目前担任《Journal of Operations Research Society of China》,《Computational Optimization and Applications》,《Mathematics of Operations Research》,《SIAM Journal on Matrix Analysis and Applications》等期刊的编委,《Numerical Algebra、Control and Optimization》共同主编(co-EiC)。他先后获得国际数学规划学会Tucker Prize Finalist奖(2009), Hellman Fellowship(2009),美国科学基金会Career奖(2009), INFORMS优化青年学者奖(2014), 2017年度长江学者, SIAM线性代数最佳论文奖(2018), 冯康科学计算奖(2021)。2022年当选美国数学会会士。