【摘要】 探讨了在强Wolfe搜索规则下,与βPRk相关的算法的收敛性。在不需要假设目标函数为凸的情况下,证明了充分下降性及算法的全局收敛性。更多还原

【Abstract】 The global convergence is considered for any conjugate gradient method of the form d1=-g1,dk=-gk+βkdk-1(k2) with any βk connected with βPRk, and with the strong Wolfe line search conditions. The sufficient descent property and the global convergence are proved for this method, without assuming the convexity of the objective function.更多还原