摘要
本文提出了一种新的算法来求解分裂可行问题,该算法在每步迭代中应用类-Arm ijo搜索来获取调整步长,然后给出了一个校正步长,避免了矩阵逆和矩阵最大特征值的计算.我们证明了该算法的全局收敛性.
In this paper,we present a new algorithm for solving the SFP,in which a predictor stepsize is first obtained by like-Armijo search and then a corrector stepsize is generated.The algorithm has the advantage that it is no need to compute the matrix inverses and the largest eigenvalue of the related matrix.The corresponding convergence properties of the algorithm are shown.
出处
《泰山学院学报》
2009年第6期9-14,共6页
Journal of Taishan University
基金
国家自然科学基金项目(10701047)
曲阜师范大学校基金资助项目(xj0625)