Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm
The Gravitational Search Algorithm (GSA) is a novel heuristic optimization method based on the law of gravity and mass interactions. It has been proven that this algorithm has good ability to search for the global optimum, but it suffers from slow searching speed in the last iterations. This work pr...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Elsevier
2012
|
| Subjects: |
| _version_ | 1796857285305171968 |
|---|---|
| author | Mirjalili, Seyed Ali Mohd. Hashim, Siti Zaiton Moradian Sardroudi, Hossein |
| author_facet | Mirjalili, Seyed Ali Mohd. Hashim, Siti Zaiton Moradian Sardroudi, Hossein |
| author_sort | Mirjalili, Seyed Ali |
| collection | ePrints |
| description | The Gravitational Search Algorithm (GSA) is a novel heuristic optimization method based on the law of gravity and mass interactions. It has been proven that this algorithm has good ability to search for the global optimum, but it suffers from slow searching speed in the last iterations. This work proposes a hybrid of Particle Swarm Optimization (PSO) and GSA to resolve the aforementioned problem. In this paper, GSA and PSOGSA are employed as new training methods for Feedforward Neural Networks (FNNs) in order to investigate the efficiencies of these algorithms in reducing the problems of trapping in local minima and the slow convergence rate of current evolutionary learning algorithms. The results are compared with a standard PSO-based learning algorithm for FNNs. The resulting accuracy of FNNs trained with PSO, GSA, and PSOGSA is also investigated. The experimental results show that PSOGSA outperforms both PSO and GSA for training FNNs in terms of converging speed and avoiding local minima. It is also proven that an FNN trained with PSOGSA has better accuracy than one trained with GSA. |
| first_indexed | 2024-03-05T18:55:23Z |
| format | Article |
| id | utm.eprints-33912 |
| institution | Universiti Teknologi Malaysia - ePrints |
| last_indexed | 2024-03-05T18:55:23Z |
| publishDate | 2012 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | utm.eprints-339122018-11-30T06:41:27Z http://eprints.utm.my/33912/ Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm Mirjalili, Seyed Ali Mohd. Hashim, Siti Zaiton Moradian Sardroudi, Hossein QA75 Electronic computers. Computer science The Gravitational Search Algorithm (GSA) is a novel heuristic optimization method based on the law of gravity and mass interactions. It has been proven that this algorithm has good ability to search for the global optimum, but it suffers from slow searching speed in the last iterations. This work proposes a hybrid of Particle Swarm Optimization (PSO) and GSA to resolve the aforementioned problem. In this paper, GSA and PSOGSA are employed as new training methods for Feedforward Neural Networks (FNNs) in order to investigate the efficiencies of these algorithms in reducing the problems of trapping in local minima and the slow convergence rate of current evolutionary learning algorithms. The results are compared with a standard PSO-based learning algorithm for FNNs. The resulting accuracy of FNNs trained with PSO, GSA, and PSOGSA is also investigated. The experimental results show that PSOGSA outperforms both PSO and GSA for training FNNs in terms of converging speed and avoiding local minima. It is also proven that an FNN trained with PSOGSA has better accuracy than one trained with GSA. Elsevier 2012-07 Article PeerReviewed Mirjalili, Seyed Ali and Mohd. Hashim, Siti Zaiton and Moradian Sardroudi, Hossein (2012) Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm. Applied Mathematics and Computation, 218 (22). pp. 11125-11137. ISSN 0096-3003 http://dx.doi.org/10.1016/j.amc.2012.04.069 DOI:10.1016/j.amc.2012.04.069 |
| spellingShingle | QA75 Electronic computers. Computer science Mirjalili, Seyed Ali Mohd. Hashim, Siti Zaiton Moradian Sardroudi, Hossein Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm |
| title | Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm |
| title_full | Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm |
| title_fullStr | Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm |
| title_full_unstemmed | Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm |
| title_short | Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm |
| title_sort | training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm |
| topic | QA75 Electronic computers. Computer science |
| work_keys_str_mv | AT mirjaliliseyedali trainingfeedforwardneuralnetworksusinghybridparticleswarmoptimizationandgravitationalsearchalgorithm AT mohdhashimsitizaiton trainingfeedforwardneuralnetworksusinghybridparticleswarmoptimizationandgravitationalsearchalgorithm AT moradiansardroudihossein trainingfeedforwardneuralnetworksusinghybridparticleswarmoptimizationandgravitationalsearchalgorithm |