A Neuro-Genetic Technique for Pruning and Optimization of ANN Weights
A novel technique for optimization of artificial neural network (ANN) weights which combines pruning and Genetic Algorithm (GA) has been proposed. The technique first defines “relevance” of initialized weights in a statistical sense by introducing a coefficient of dominance for each weight and subse...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2019-01-01
|
| Series: | Applied Artificial Intelligence |
| Online Access: | http://dx.doi.org/10.1080/08839514.2018.1525524 |
| _version_ | 1827817599918931968 |
|---|---|
| author | Sakshi Sakshi Ravi Kumar |
| author_facet | Sakshi Sakshi Ravi Kumar |
| author_sort | Sakshi Sakshi |
| collection | DOAJ |
| description | A novel technique for optimization of artificial neural network (ANN) weights which combines pruning and Genetic Algorithm (GA) has been proposed. The technique first defines “relevance” of initialized weights in a statistical sense by introducing a coefficient of dominance for each weight and subsequently employing the concept of complexity penalty. Based upon complexity penalty for each weight, candidate solutions are initialized to participate in the Genetic optimization. The GA stage employs mean square error as the fitness function which is evaluated once for all candidate solutions by running the forward pass of backpropagation. Subsequent reproduction cycles generate fitter individuals and the GA is terminated after a small number of cycles. It has been observed that ANNs trained with GA optimized weights exhibit higher convergence, lower execution time, and higher success rate in the test phase. Furthermore, the proposed technique yields substantial reduction in computational resources. |
| first_indexed | 2024-03-12T00:36:32Z |
| format | Article |
| id | doaj.art-c722f66c84b247bbba79e555980f6998 |
| institution | Directory Open Access Journal |
| issn | 0883-9514 1087-6545 |
| language | English |
| last_indexed | 2024-03-12T00:36:32Z |
| publishDate | 2019-01-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Applied Artificial Intelligence |
| spelling | doaj.art-c722f66c84b247bbba79e555980f69982023-09-15T09:33:56ZengTaylor & Francis GroupApplied Artificial Intelligence0883-95141087-65452019-01-0133112610.1080/08839514.2018.15255241525524A Neuro-Genetic Technique for Pruning and Optimization of ANN WeightsSakshi Sakshi0Ravi Kumar1Thapar UniversityThapar UniversityA novel technique for optimization of artificial neural network (ANN) weights which combines pruning and Genetic Algorithm (GA) has been proposed. The technique first defines “relevance” of initialized weights in a statistical sense by introducing a coefficient of dominance for each weight and subsequently employing the concept of complexity penalty. Based upon complexity penalty for each weight, candidate solutions are initialized to participate in the Genetic optimization. The GA stage employs mean square error as the fitness function which is evaluated once for all candidate solutions by running the forward pass of backpropagation. Subsequent reproduction cycles generate fitter individuals and the GA is terminated after a small number of cycles. It has been observed that ANNs trained with GA optimized weights exhibit higher convergence, lower execution time, and higher success rate in the test phase. Furthermore, the proposed technique yields substantial reduction in computational resources.http://dx.doi.org/10.1080/08839514.2018.1525524 |
| spellingShingle | Sakshi Sakshi Ravi Kumar A Neuro-Genetic Technique for Pruning and Optimization of ANN Weights Applied Artificial Intelligence |
| title | A Neuro-Genetic Technique for Pruning and Optimization of ANN Weights |
| title_full | A Neuro-Genetic Technique for Pruning and Optimization of ANN Weights |
| title_fullStr | A Neuro-Genetic Technique for Pruning and Optimization of ANN Weights |
| title_full_unstemmed | A Neuro-Genetic Technique for Pruning and Optimization of ANN Weights |
| title_short | A Neuro-Genetic Technique for Pruning and Optimization of ANN Weights |
| title_sort | neuro genetic technique for pruning and optimization of ann weights |
| url | http://dx.doi.org/10.1080/08839514.2018.1525524 |
| work_keys_str_mv | AT sakshisakshi aneurogenetictechniqueforpruningandoptimizationofannweights AT ravikumar aneurogenetictechniqueforpruningandoptimizationofannweights AT sakshisakshi neurogenetictechniqueforpruningandoptimizationofannweights AT ravikumar neurogenetictechniqueforpruningandoptimizationofannweights |