Abstract
Earlier attempts are made to design Hopfield type neural network architecture for object extraction using Genetic Algorithms (GAs). Energy value of the neural network was taken as the index of fitness of the GA. In the present article fuzzy logic reasoning is incorporated into this Neuro-GA hybrid framework to remove some of the drawbacks of earlier attempts. Here, GAs have been used to evolve Hopfield type optimum neural network architecture for object background classification. Each chromosome of the GA represents an architecture. The output status of the neurons at the converged state of the network is viewed as a fuzzy set and measure of fuzziness of this set is taken as a measure of fitness of the chromosome. The best chromosome of the final generation represents the optimum network configuration. When the input images are less noisy, the evolved networks are found to have less (compared to the corresponding energy based objective evaluation) connectivity for providing comparable outputs.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Mitchell, M.: An Introduction to Genetic Algorithms. The MIT Press, Cambridge, Massachusetts (1996)
Whitley, D., Starkweather, T., Bogart, C.: Genetic algorithms and neural networks: Optimizing connections and connectivity. Parallel Computing. 14 (1990) 347–361
Pal, S. K., Bhandari, D.:Selection of optimal set of weights in a layered network using genetic algorithms. Information Sciences. 80 (1994) 213–234
De, S., Ghosh, A., Pal, S. K.:An application of genetic algorithms to evolve Hopfield type optimum network architectures for object extraction. Proceedings 1995 IEEE International Conference on Evolutionary Computation, ICEC’95, Perth, Western Australia. 1 (1995) 504–508
Hopfield, J. J.: Neural network and physical systems with emergent collective computational abilities. Proceedings National Academy of Science, USA. 79 (1982) 2554–2558
Ghosh, A., Pal, N. R., Pal, S. K.: Object background classification using Hopfield type neural network. International Journal of Pattern recognition and Artificial Intelligence. 6 (1992) 989–1008
Pal, S. K., De, S., Ghosh, A.: Designing Hopfield type networks using genetic algorithms and its comparison with simulated annealing. International Journal of Pattern Recognition and Artificial Intelligence. 11 (1997) 447–461
Ghosh, A.: Use of fuzziness measures in layered networks for object extraction: a generalization. Fuzzy Sets and Systems. 72 (1995) 331–348
Klir, G. J., Folger, T.: Fuzzy Sets, Uncertainty and Information. Prentice-Hall, Englewood, Cliffs (1988)
Davis, L. (ed.): Handbook of Genetic Algorithms. Van Nostrand Reinhold, New York (1991)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ghosh, S., Ghosh, A. (2002). A GA-FUZZY Approach to Evolve Hopfield Type Optimum Networks for Object Extraction. In: Pal, N.R., Sugeno, M. (eds) Advances in Soft Computing — AFSS 2002. AFSS 2002. Lecture Notes in Computer Science(), vol 2275. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45631-7_60
Download citation
DOI: https://doi.org/10.1007/3-540-45631-7_60
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-43150-3
Online ISBN: 978-3-540-45631-5
eBook Packages: Springer Book Archive