摘要
给出了一个正定二次规划的对偶算法 .算法把原问题分解为一系列子问题 ,在保持原问题的 Wolfe对偶可行的前提下 ,通过迭代计算 ,由这一系列子问题的最优解向原问题的最优解逼近 .同时给出了算法的有限收敛性 .
A dual method for solving positive definite quadratic programming is given. The original problem is divided into a series of subproblems in the method. In the condition of Wolfe dual problem is feasible, the optimum solution is obtained by solving the subproblems through iterations. Finite termination also is proved.
出处
《纯粹数学与应用数学》
CSCD
2000年第4期15-20,共6页
Pure and Applied Mathematics
基金
西北工业大学"双新计划"资助
关键词
正定二次规划
Wolfe对偶
有效约束
positive definite quadratic programming
Wolfe dual problem
active constraint
