A New Adaptive Weighted Hybrid Conjugate Gradient Method for Unconstrained Optimization Problems
Journal of Decision Science and Optimization · 02.06.2026 · 2 (2) · 97-112
DOI: 10.55578/jdso.2606.007 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Decision Science and Optimization, online erschienen am 02.06.2026. Verfasst von Yunchol Jong und 2 weiteren Autor:innen.
Abstract (Originalsprache)
In this paper, a new adaptive weighted hybrid conjugate gradient (AWHCG) method with backtracking line search is proposed to solve unconstrained optimization problems. Our method is based on the weighted hybridization of HS (Hestenes-Stiefel) and FR (Fletcher-Reeves) conjugate gradient methods. In contrast with the existing hybrid conjugate gradient methods, our AWHCG method updates adaptively the coefficient of gradient of objective function to be negative at each iteration so that our current search direction is always a combination of the anti-gradient of the objective function and the preceding direction. Under some reasonable assumptions, the global convergence of the proposed algorithm is established. Some preliminary numerical experiments show that the AWHCG algorithm is competitive.
Zitieren
APA
Jong, Y., Kim, Y., & Ri, D. (2026). A New Adaptive Weighted Hybrid Conjugate Gradient Method for Unconstrained Optimization Problems. Journal of Decision Science and Optimization, 2(2), 97-112. https://doi.org/10.55578/jdso.2606.007
Vancouver
Jong Y, Kim Y, Ri D. A New Adaptive Weighted Hybrid Conjugate Gradient Method for Unconstrained Optimization Problems. Journal of Decision Science and Optimization. 2026;2(2):97-112. doi:10.55578/jdso.2606.007
BibTeX
@article{Jong2026,
author = {Jong, Yunchol and Kim, Yongjin and Ri, Daesong},
title = {A New Adaptive Weighted Hybrid Conjugate Gradient Method for Unconstrained Optimization Problems},
journal = {Journal of Decision Science and Optimization},
year = {2026},
volume = {2},
number = {2},
pages = {97--112},
doi = {10.55578/jdso.2606.007}
}
EUVIMED