求解非凸函数极小问题的一类改进BFGS算法的收敛性分析

Convergence Analysis of a Class of Improved BFGS Algorithm on Nonconvex Funtions Minimal Problems

提出了一类求解无约束最优化问题的改进BFGS算法,主要讨论了该算法在Wolfe搜索下的全局收敛性.数值试验结果表明新算法是有效的.

In the paper,a class of improved BFGS algorithm is presented for unconstained optimization problems.The global convergence is proved with Wolfe line search under some conditions and numerical results show that the new algorithm is effective.