A decomposition algorithm for a class of constrained nonsmooth convex optimization problems
In this paper we propose a decomposition algorithm for solving a class of constrained nonsmooth convex optimization problems which constraints are another nonsmooth minimization problems.The original problem is approximately replaced by minimizing a parameterized family of functions instead of penalty functions.The family of these functions is the sum of two convex functions and we decompose the parameterized problem to two approximate subproblems into exploit the substructures of each function.It is shown that the accumulation points of iterative sequence belong to the solution set of the original problem.Numerical experiments validate the theoretical convergence analysis and illustrate the implementation of the decomposition algorithm.
nonlinear programming convex programming nonsmooth optimization proximal point methods alternating linearization algorithms
Dan Li Shuang Chen Fan-Yun Meng
Information and Engineering College Dalian University Dalian,China School of Mathematical Sciences Dalian University of Technology Dalian,China
国际会议
重庆
英文
685-690
2016-03-20（万方平台首次上网日期，不代表论文的发表时间）