Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training
In this study, a machine learning approach based on the unsupervised version of wavelet neural networks (WNNs) is used to solve two-dimensional elliptic partial differential equations (PDEs). The design of the WNNs must be judiciously addressed, particularly, the adopted training algorithm, since it...
Saved in:
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2020
|
Subjects: | |
Online Access: | http://eprints.uthm.edu.my/6650/1/AJ%202020%20%28411%29.pdf http://eprints.uthm.edu.my/6650/ https://doi.org/10.1016/j.asoc.2020.106518 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.uthm.eprints.6650 |
---|---|
record_format |
eprints |
spelling |
my.uthm.eprints.66502022-03-14T01:31:40Z http://eprints.uthm.edu.my/6650/ Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training Lee, Sen Tan Zainuddin, Zarita Ong, Pauline TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television In this study, a machine learning approach based on the unsupervised version of wavelet neural networks (WNNs) is used to solve two-dimensional elliptic partial differential equations (PDEs). The design of the WNNs must be judiciously addressed, particularly, the adopted training algorithm, since it greatly influences the generalization performance and the convergence rate of WNNs. Although the gradient information of the commonly used gradient descent training algorithm in WNNs may direct the search to optimal weight solutions that minimize the error function, the learning process is slow due to the complex calculation of the partial derivatives. To date, on account of the derivative free characteristic and adaptability to respond to the complex dynamic changes of the interdependencies, numerous studies explored the potential benefit of integrating a meta-heuristic algorithm as the training algorithm of WNNs, where encouraging results are achieved. In this paper, an improved butterfly optimization algorithm (IBOA) is proposed and subsequently integrated into the training process of the WNNs. To evaluate the performance of the proposed IBOA training method, the obtained results are compared to the results of the momentum backpropagation (MBP), the particle swarm optimization (PSO) and the standard butterfly optimization algorithm (BOA) training methods. Statistical analyses of the results based on a sufficient number of independent runs validate the effectiveness of the proposed method in terms of accuracy, robustness and convergence. Elsevier 2020 Article PeerReviewed text en http://eprints.uthm.edu.my/6650/1/AJ%202020%20%28411%29.pdf Lee, Sen Tan and Zainuddin, Zarita and Ong, Pauline (2020) Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training. Applied Soft Computing Journal, 95. pp. 1-11. ISSN 1568-4946 https://doi.org/10.1016/j.asoc.2020.106518 |
institution |
Universiti Tun Hussein Onn Malaysia |
building |
UTHM Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Tun Hussein Onn Malaysia |
content_source |
UTHM Institutional Repository |
url_provider |
http://eprints.uthm.edu.my/ |
language |
English |
topic |
TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television |
spellingShingle |
TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television Lee, Sen Tan Zainuddin, Zarita Ong, Pauline Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training |
description |
In this study, a machine learning approach based on the unsupervised version of wavelet neural networks (WNNs) is used to solve two-dimensional elliptic partial differential equations (PDEs). The design of the WNNs must be judiciously addressed, particularly, the adopted training algorithm, since it greatly influences the generalization performance and the convergence rate of WNNs. Although the gradient information of the commonly used gradient descent training algorithm in WNNs may direct the search to optimal weight solutions that minimize the error function, the learning process is slow due to the complex calculation of the partial derivatives. To date, on account of the derivative free characteristic and adaptability to respond to the complex dynamic changes of the interdependencies, numerous studies explored the potential benefit of integrating a meta-heuristic algorithm as the training algorithm of WNNs, where encouraging results are achieved. In this paper, an improved butterfly optimization algorithm (IBOA) is proposed and subsequently integrated into the training process of the WNNs. To evaluate the performance of the proposed IBOA training method, the obtained results are compared to the results of the momentum backpropagation (MBP), the particle swarm optimization (PSO) and the standard butterfly optimization algorithm (BOA) training methods. Statistical analyses of the results based on a sufficient number of independent runs validate the effectiveness of the proposed method in terms of accuracy, robustness and convergence. |
format |
Article |
author |
Lee, Sen Tan Zainuddin, Zarita Ong, Pauline |
author_facet |
Lee, Sen Tan Zainuddin, Zarita Ong, Pauline |
author_sort |
Lee, Sen Tan |
title |
Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training |
title_short |
Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training |
title_full |
Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training |
title_fullStr |
Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training |
title_full_unstemmed |
Wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training |
title_sort |
wavelet neural networks based solutions for elliptic partial differential equations with improved butterfly optimization algorithm training |
publisher |
Elsevier |
publishDate |
2020 |
url |
http://eprints.uthm.edu.my/6650/1/AJ%202020%20%28411%29.pdf http://eprints.uthm.edu.my/6650/ https://doi.org/10.1016/j.asoc.2020.106518 |
_version_ |
1738581517721927680 |
score |
13.160551 |