Basri, S.M.M. and Nawi, N.M. and Mamat, M. and Hamid, N.A. (2018) RMIL/AG: A new class of nonlinear conjugate gradient for training back propagation algorithm. In: 3rd International Conference on Soft Computing and Data Mining, 06-08 Feb 2018, Johor, Malaysia.
![[img]](http://eprints.unisza.edu.my/1728/1/FH03-FIK-18-13578.jpg) |
Image
FH03-FIK-18-13578.jpg Restricted to Registered users only Download (155kB) |
![[img]](http://eprints.unisza.edu.my/style/images/fileicons/text.png) |
Text
FH03-FIK-18-18362.pdf Restricted to Registered users only Download (341kB) |
Abstract
The conventional back propagation (BP) algorithm is generally known for some disadvantages, such as slow training, easy to getting trapped into local minima and being sensitive to the initial weights and bias. This paper introduced a new class of efficient second order conjugate gradient (CG) for training BP called Rivaie, Mustafa, Ismail and Leong (RMIL)/AG. The RMIL uses the value of adaptive gain parameter in the activation function to modify the gradient based search direction. The efficiency of the proposed method is verified by means of simulation on four classification problems. The results show that the computational efficiency of the proposed method was better than the conventional BP algorithm.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Uncontrolled Keywords: | second order method, back-propagation, conjugate gradient, search direction, classification |
| Subjects: | Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science T Technology > T Technology (General) |
| Divisions: | Faculty of Informatics & Computing |
| Depositing User: | Muhammad Akmal Azhar |
| Date Deposited: | 22 Nov 2020 03:36 |
| Last Modified: | 22 Nov 2020 03:36 |
| URI: | http://eprints.unisza.edu.my/id/eprint/1728 |
Actions (login required)
 |
View Item |
