主讲人:Professor Xu, Hong-Kun
时间:2019年12月3日13:00
地点:数学系332室
举办单位:数理学院
主讲人介绍:南非国家科学院院士,发展中国家科学院院士
内容介绍:Subgradient methods, introduced by Shor and developed by Albert, Iusem,Nesterov, Polyak, Soloov, and many others, are used to solve nondifferentiableoptimization problems. The major differences from the gradient descent methods(or projection-gradient methods) for differentiable optimization problems lie in theselection manners of the step-sizes. For instance, constant step-sizes for differen-tiable objective functions no longer work for nondifferentiable objective functions;for the latter case, diminishing step-sizes must however be adopted.In this talk, we will first review some existing projected subgradient methodsand the main purpose is to discuss weak and strong convergence of projected sub-gradient methods in an infinite-dimensional Hilbert space. Some novel approachesfor strong convergence analysis of projected subgradient methods will particularlybe presented.