.png){kind=icon}
.png){kind=icon}
Return
Two Modified Hybrid Conjugate Gradient Methods for Nonconvex Vector Optimization
.png){kind=icon}
PRE
AI
DOI：10.11650/tjm/251104.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this paper, we develop two modified hybrid conjugate gradient methods for vector optimization involving three modified conjugate parameters-Fletcher-Reeves (FR) type, Conjugate Descent (CD) type, and Polak-Ribi & eacute;re-Polyak (PRP) type. The modified FR and CD parameters guarantee the sufficient descent condition without special line searches, while the modified PRP offers strong numerical performance. Leveraging these strengths, we propose two modified hybrid conjugate gradient methods that combine the modified FR/CD conjugate parameters with the PRP conjugate parameter. Global convergence results of the two hybrid conjugate gradient methods are established under the strong Wolfe line search strategy without convexity assumptions. Numerical results show the effectiveness and advantages of the proposed methods.
Keywords:
global convergence
hybrid conjugate gradient method
sufficient descent condition
vector optimization
Journal
T
IF:
0.6
Papers:
37
Citations:
0