Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout Regularization
Swarm intelligence techniques have been created to respond to theoretical and practical global optimization problems. This paper puts forward an enhanced version of the firefly algorithm that corrects the acknowledged drawbacks of the original method, by an explicit exploration mechanism and a chaot...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-10-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/9/21/2705 |
| _version_ | 1797512050753142784 |
|---|---|
| author | Nebojsa Bacanin Ruxandra Stoean Miodrag Zivkovic Aleksandar Petrovic Tarik A. Rashid Timea Bezdan |
| author_facet | Nebojsa Bacanin Ruxandra Stoean Miodrag Zivkovic Aleksandar Petrovic Tarik A. Rashid Timea Bezdan |
| author_sort | Nebojsa Bacanin |
| collection | DOAJ |
| description | Swarm intelligence techniques have been created to respond to theoretical and practical global optimization problems. This paper puts forward an enhanced version of the firefly algorithm that corrects the acknowledged drawbacks of the original method, by an explicit exploration mechanism and a chaotic local search strategy. The resulting augmented approach was theoretically tested on two sets of bound-constrained benchmark functions from the CEC suites and practically validated for automatically selecting the optimal dropout rate for the regularization of deep neural networks. Despite their successful applications in a wide spectrum of different fields, one important problem that deep learning algorithms face is overfitting. The traditional way of preventing overfitting is to apply regularization; the first option in this sense is the choice of an adequate value for the dropout parameter. In order to demonstrate its ability in finding an optimal dropout rate, the boosted version of the firefly algorithm has been validated for the deep learning subfield of convolutional neural networks, with respect to five standard benchmark datasets for image processing: MNIST, Fashion-MNIST, Semeion, USPS and CIFAR-10. The performance of the proposed approach in both types of experiments was compared with other recent state-of-the-art methods. To prove that there are significant improvements in results, statistical tests were conducted. Based on the experimental data, it can be concluded that the proposed algorithm clearly outperforms other approaches. |
| first_indexed | 2024-03-10T05:56:26Z |
| format | Article |
| id | doaj.art-052c5d230f964fffa39742d8afb6a1f7 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-10T05:56:26Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-052c5d230f964fffa39742d8afb6a1f72023-11-22T21:17:38ZengMDPI AGMathematics2227-73902021-10-01921270510.3390/math9212705Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout RegularizationNebojsa Bacanin0Ruxandra Stoean1Miodrag Zivkovic2Aleksandar Petrovic3Tarik A. Rashid4Timea Bezdan5Faculty of Informatics and Computing, Singidunum University, Danijelova 32, 11000 Belgrade, SerbiaRomanian Institute of Science and Technology, Str. Virgil Fulicea 3, 400022 Cluj-Napoca, RomaniaFaculty of Informatics and Computing, Singidunum University, Danijelova 32, 11000 Belgrade, SerbiaFaculty of Informatics and Computing, Singidunum University, Danijelova 32, 11000 Belgrade, SerbiaComputer Science and Engineering, University of Kurdistan Hewler, 30 Meter Avenue, Erbil 44001, IraqFaculty of Informatics and Computing, Singidunum University, Danijelova 32, 11000 Belgrade, SerbiaSwarm intelligence techniques have been created to respond to theoretical and practical global optimization problems. This paper puts forward an enhanced version of the firefly algorithm that corrects the acknowledged drawbacks of the original method, by an explicit exploration mechanism and a chaotic local search strategy. The resulting augmented approach was theoretically tested on two sets of bound-constrained benchmark functions from the CEC suites and practically validated for automatically selecting the optimal dropout rate for the regularization of deep neural networks. Despite their successful applications in a wide spectrum of different fields, one important problem that deep learning algorithms face is overfitting. The traditional way of preventing overfitting is to apply regularization; the first option in this sense is the choice of an adequate value for the dropout parameter. In order to demonstrate its ability in finding an optimal dropout rate, the boosted version of the firefly algorithm has been validated for the deep learning subfield of convolutional neural networks, with respect to five standard benchmark datasets for image processing: MNIST, Fashion-MNIST, Semeion, USPS and CIFAR-10. The performance of the proposed approach in both types of experiments was compared with other recent state-of-the-art methods. To prove that there are significant improvements in results, statistical tests were conducted. Based on the experimental data, it can be concluded that the proposed algorithm clearly outperforms other approaches.https://www.mdpi.com/2227-7390/9/21/2705convolutional neural networksdropoutregularizationmetaheuristicsswarm intelligenceoptimization |
| spellingShingle | Nebojsa Bacanin Ruxandra Stoean Miodrag Zivkovic Aleksandar Petrovic Tarik A. Rashid Timea Bezdan Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout Regularization Mathematics convolutional neural networks dropout regularization metaheuristics swarm intelligence optimization |
| title | Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout Regularization |
| title_full | Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout Regularization |
| title_fullStr | Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout Regularization |
| title_full_unstemmed | Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout Regularization |
| title_short | Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout Regularization |
| title_sort | performance of a novel chaotic firefly algorithm with enhanced exploration for tackling global optimization problems application for dropout regularization |
| topic | convolutional neural networks dropout regularization metaheuristics swarm intelligence optimization |
| url | https://www.mdpi.com/2227-7390/9/21/2705 |
| work_keys_str_mv | AT nebojsabacanin performanceofanovelchaoticfireflyalgorithmwithenhancedexplorationfortacklingglobaloptimizationproblemsapplicationfordropoutregularization AT ruxandrastoean performanceofanovelchaoticfireflyalgorithmwithenhancedexplorationfortacklingglobaloptimizationproblemsapplicationfordropoutregularization AT miodragzivkovic performanceofanovelchaoticfireflyalgorithmwithenhancedexplorationfortacklingglobaloptimizationproblemsapplicationfordropoutregularization AT aleksandarpetrovic performanceofanovelchaoticfireflyalgorithmwithenhancedexplorationfortacklingglobaloptimizationproblemsapplicationfordropoutregularization AT tarikarashid performanceofanovelchaoticfireflyalgorithmwithenhancedexplorationfortacklingglobaloptimizationproblemsapplicationfordropoutregularization AT timeabezdan performanceofanovelchaoticfireflyalgorithmwithenhancedexplorationfortacklingglobaloptimizationproblemsapplicationfordropoutregularization |