改进的无约束化的BFGS算法

改进的无约束化的BFGS算法 AN IMPROVED BFGS METHOD FOR UNCONSTRAINED MINIMIZATION 费浦生 Fei Pusheng 1 first-author 陈忠 Chen Zhong 1 武汉大学数学系 武汉大学数学系 Department of Mathematics, Wuhan University Department of Mathematics, Wuhan University 提出了一种改进的ＢＦＧＳ算法，该算法的基本思想是采用多步迭代改变一次ＢＦＧＳ算法的修正公式；在假设ｆ（ｘ）是二阶连续可微且一致凸的条件下，证明了该算法的超线性收敛性，最后给出了该算法的数值实验结果． We present an improved BFGS method,if we assume the objective function iS twice continuouSly differential and uniformly convex,we prove the iterates converge globally and superlinearly to the solution. Finally we show the numerical resultS of this algorithm. 改进的BFGS算法 无约束优化 超线性收敛性 BFGS method superlinear convergence unconstrained minimization O224 国家自然科学基金(61672102) 1994-03-01 2021-04-01 3