An improved back propagation leaning algorithm using second order methods with gain parameter
Back Propagation (BP) algorithm is one of the oldest learning techniques used by Artificial Neural Networks (ANN). It has successfully been implemented in various practical problems. However, the algorithm still faces some drawbacks such as getting easily stuck at local minima and needs longer time...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Penerbit UTHM
2018
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/4558/ http://eprints.uthm.edu.my/4558/1/AJ%202018%20%28791%29%20An%20improved%20back%20propagation%20leaning%20algorithm%20using%20second%20order%20methods%20with%20gain%20parameter.pdf |
| _version_ | 1848888319511887872 |
|---|---|
| author | Mohd Nawi, Nazri Mohamed Saufi, Noor Haliza Budiyono, Avon Abdul Hamid, Noorhamreeza Rehman Gillani, Syed Muhammad Zubair Ramli, Azizul Azhar |
| author_facet | Mohd Nawi, Nazri Mohamed Saufi, Noor Haliza Budiyono, Avon Abdul Hamid, Noorhamreeza Rehman Gillani, Syed Muhammad Zubair Ramli, Azizul Azhar |
| author_sort | Mohd Nawi, Nazri |
| building | UTHM Institutional Repository |
| collection | Online Access |
| description | Back Propagation (BP) algorithm is one of the oldest learning techniques used by Artificial Neural Networks (ANN). It has successfully been implemented in various practical problems. However, the algorithm still faces some drawbacks such as getting easily stuck at local minima and needs longer time to converge on an acceptable solution. Recently, the introduction of Second Order Methods has shown a significant improvement on the learning in BP but it still has some drawbacks such as slow convergence and complexity. To overcome these limitations, this research proposed a modified approach for BP by introducing the Conjugate Gradient and QuasiNewton which were Second Order methods together with ‘gain’ parameter. The performances of the proposed approach is evaluated in terms of lowest number of epochs, lowest CPU time and highest accuracy on five benchmark classification datasets such as Glass, Horse, 7Bit Parity, Indian Liver Patient and Lung Cancer. The results show that the proposed Second Order methods with ‘gain’ performed better than the BP algorithm. |
| first_indexed | 2025-11-15T20:08:24Z |
| format | Article |
| id | uthm-4558 |
| institution | Universiti Tun Hussein Onn Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T20:08:24Z |
| publishDate | 2018 |
| publisher | Penerbit UTHM |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | uthm-45582021-12-07T07:17:49Z http://eprints.uthm.edu.my/4558/ An improved back propagation leaning algorithm using second order methods with gain parameter Mohd Nawi, Nazri Mohamed Saufi, Noor Haliza Budiyono, Avon Abdul Hamid, Noorhamreeza Rehman Gillani, Syed Muhammad Zubair Ramli, Azizul Azhar TA168 Systems engineering TS155-194 Production management. Operations management Back Propagation (BP) algorithm is one of the oldest learning techniques used by Artificial Neural Networks (ANN). It has successfully been implemented in various practical problems. However, the algorithm still faces some drawbacks such as getting easily stuck at local minima and needs longer time to converge on an acceptable solution. Recently, the introduction of Second Order Methods has shown a significant improvement on the learning in BP but it still has some drawbacks such as slow convergence and complexity. To overcome these limitations, this research proposed a modified approach for BP by introducing the Conjugate Gradient and QuasiNewton which were Second Order methods together with ‘gain’ parameter. The performances of the proposed approach is evaluated in terms of lowest number of epochs, lowest CPU time and highest accuracy on five benchmark classification datasets such as Glass, Horse, 7Bit Parity, Indian Liver Patient and Lung Cancer. The results show that the proposed Second Order methods with ‘gain’ performed better than the BP algorithm. Penerbit UTHM 2018 Article PeerReviewed text en http://eprints.uthm.edu.my/4558/1/AJ%202018%20%28791%29%20An%20improved%20back%20propagation%20leaning%20algorithm%20using%20second%20order%20methods%20with%20gain%20parameter.pdf Mohd Nawi, Nazri and Mohamed Saufi, Noor Haliza and Budiyono, Avon and Abdul Hamid, Noorhamreeza and Rehman Gillani, Syed Muhammad Zubair and Ramli, Azizul Azhar (2018) An improved back propagation leaning algorithm using second order methods with gain parameter. International Journal of Integrated Engineering, 10 (6). pp. 11-18. ISSN 2229-838X |
| spellingShingle | TA168 Systems engineering TS155-194 Production management. Operations management Mohd Nawi, Nazri Mohamed Saufi, Noor Haliza Budiyono, Avon Abdul Hamid, Noorhamreeza Rehman Gillani, Syed Muhammad Zubair Ramli, Azizul Azhar An improved back propagation leaning algorithm using second order methods with gain parameter |
| title | An improved back propagation leaning algorithm using second order methods with gain parameter |
| title_full | An improved back propagation leaning algorithm using second order methods with gain parameter |
| title_fullStr | An improved back propagation leaning algorithm using second order methods with gain parameter |
| title_full_unstemmed | An improved back propagation leaning algorithm using second order methods with gain parameter |
| title_short | An improved back propagation leaning algorithm using second order methods with gain parameter |
| title_sort | improved back propagation leaning algorithm using second order methods with gain parameter |
| topic | TA168 Systems engineering TS155-194 Production management. Operations management |
| url | http://eprints.uthm.edu.my/4558/ http://eprints.uthm.edu.my/4558/1/AJ%202018%20%28791%29%20An%20improved%20back%20propagation%20leaning%20algorithm%20using%20second%20order%20methods%20with%20gain%20parameter.pdf |