A modified conjugate gradient coefficient under exact line search for unconstrained optimization
The conjugate gradient (CG) method is a well-known solver for large-scale unconstrained optimization problems. In this paper, a modified CG method based on AMR* and CD method is presented. The resulting algorithm for the new CG method is proved to be globally convergent under exact line search both...
| Published in: | International Journal of Engineering and Technology(UAE) |
|---|---|
| Main Author: | |
| Format: | Article |
| Language: | English |
| Published: |
Science Publishing Corporation Inc
2018
|
| Online Access: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85076105320&partnerID=40&md5=74856ca33d2f1cec66a970563a029a62 |
| Summary: | The conjugate gradient (CG) method is a well-known solver for large-scale unconstrained optimization problems. In this paper, a modified CG method based on AMR* and CD method is presented. The resulting algorithm for the new CG method is proved to be globally convergent under exact line search both under some mild conditions. Comparisons of numerical performance are made involving the new method and four other CG methods. The results show that the proposed method is more efficient. © 2018 Authors. |
|---|---|
| ISSN: | 2227524X |