摘要
目前对一般无约束最优化问题的研究主要是以拟牛顿算法为主,并且随着研究的深入,各种基于修改原始拟牛顿方程的新拟牛顿算法取得了较好的成果。文章主要对几种改进后的算法做个综述。
At present there is no constraint optimized nesting for the general problem of research mainly to planning, and with Newton＇s method of study, all based on the original proposed amendment Newton of the equation is a new algorithm for better results. This article mainly for several improved algorithm for a review.
出处
《大众科技》
2011年第11期3-4,10,共3页
Popular Science & Technology
关键词
BFGS算法
拟牛顿算法
全局收敛性
超线性收敛
BFGS algoritm
quasi-Newton method
global convergence
superlinear convergence
