Back to Search
Start Over
An adaptive fractional-order BP neural network based on extremal optimization for handwritten digits recognition.
- Source :
-
Neurocomputing . May2020, Vol. 391, p260-272. 13p. - Publication Year :
- 2020
-
Abstract
- The optimal generation of initial connection weight parameters and dynamic updating strategies of connection weights are critical for adjusting the performance of back-propagation (BP) neural networks. This paper presents an adaptive fractional-order BP neural network abbreviated as PEO-FOBP for handwritten digit recognition problems by combining a competitive evolutionary algorithm called population extremal optimization and a fractional-order gradient descent learning mechanism. Population extremal optimization is introduced to optimize a large number of initial connection weight parameters and fractional-order gradient descent learning mechanism is designed to update these connection weight parameters adaptively during the evolutionary process of fractional-order BP neural network. The extensive experimental results for a well-known MNIST handwritten digits dataset have demonstrated that the proposed PEO-FOBP outperforms the original fractional-order BP neural network and the traditional integer-order BP neural network in terms of training and testing accuracies. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 09252312
- Volume :
- 391
- Database :
- Academic Search Index
- Journal :
- Neurocomputing
- Publication Type :
- Academic Journal
- Accession number :
- 143059970
- Full Text :
- https://doi.org/10.1016/j.neucom.2018.10.090