Automatic Differentiation Based for Particle Swarm Optimization Steepest Descent Direction
| Main Authors: | Thobirin, Aris, Tri Riyadi Yanto, Iwan |
|---|---|
| Format: | Article PeerReviewed Book |
| Bahasa: | eng |
| Terbitan: |
http://ijain.org/index.php/IJAIN/article/view/29
, 2018
|
| Subjects: | |
| Online Access: |
http://eprints.uad.ac.id/11644/1/Automatic%20Differentiation%20Based%20for%20Particle%20Swarm%20Optimization%20Steepest%20Descent%20Direction.pdf http://eprints.uad.ac.id/11644/ |
Daftar Isi:
- Particle swam optimization (PSO) is one of the most effective optimization methods to find the global optimum point. In other hand, the descent direction (DD) is the gradient based method that has the local search capability. The combination of both methods is promising and interesting to get the method with effective global search capability and efficient local search capability. However, In many application, it is difficult or impossible to obtain the gradient exactly of an objective function. In this paper, we propose Automatic differentiation (AD) based for PSODD. We compare our methods on benchmark function. The results shown that the combination methods give us a powerful tool to find the solution.