向量优化的Dai-Kou共轭梯度法
首发时间:2024-03-04
摘要:向量优化是最优化邻域中的主要研究方向之一,共轭梯度法是求解无约束优化问题的一类重要的一阶算法,近年来被推广应用于向量优化问题。本文首先简要叙述了一些向量优化的共轭梯度法的理论知识与技术手段;其次将Dai-Kou共轭梯度法由标量形式推广到向量形式,并通过自适应调整策略完善算法,提出了有效的向量优化的四种不同参数下的DK共轭梯度法,并且分别讨论四种共轭参数下的充分下降性,证明其全局收敛性的结果;最后,通过一系列的数值实验,与现有向量优化的共轭梯度法进行比较,表明本文提出的DK共轭梯度法的有效性。
关键词: 共轭梯度法 向量优化 无约束优化 全局收敛 wolfe条件
For information in English, please click here
Vector optimization of Dai-Kou conjugate gradient method
Abstract:Vector optimization is one of the main research directions in the optimization neighborhood. Conjugate gradient method is an important first-order algorithm for solving unconstrained optimization problems, which has been extended to vector optimization problems in recent years. Firstly, this paper briefly describes some theoretical knowledge and technical means of conjugate gradient method for vector optimization. Secondly, the Dai-Kou conjugate gradient method is extended from scalar form to vector form, and the algorithm is improved by adaptive adjustment strategy. The effective vector optimization DK conjugate gradient method under four different parameters is proposed, and the sufficient descent under four conjugate parameters is discussed respectively, and the result of global convergence is proved. Finally, through a series of numerical experiments, the effectiveness of DK conjugate gradient method proposed in this paper is demonstrated by comparison with the existing conjugate gradient method of vector optimization.
Keywords: Conjugate gradient method Vector optimization Unconstrained optimization Global convergence wolfe condition
基金:
引用
No.****
同行评议
勘误表
向量优化的Dai-Kou共轭梯度法
评论
全部评论