凸规划的新算法

A NEW ALGORITHM FOR CONVEX PROGRAMMING

对一般凸目标函数和一般凸集约束的凸规划问题新解法进行探讨,它是线性规划一种新算法的扩展和改进.此算法的基本思想是在规划问题的可行域中由所建的一个切割面到另一个切割面的不断推进来求取最优的.文章对目标函数是二次的且约束是一般凸集和二次目标函数且约束是线性的情形,给出了更简单的算法.

This paper is a discussion of new algorithm for convex objective function and convex programming with constraints.The algorithm is an extension and improvement for a new linear programming.The idea of this algorithm is to find the optimal solution in the feasible region by an iterative step from one basic standard hyperplane to another.A simple method for quadratic objective function with convex constaints and with linear constaints is given.