HYBRID OPTIMIZATION TECHNIQUE FOR ARTIFICIAL NEURAL NETWORKS DESIGN

Cleber Zanchettin, Teresa B. Ludermir

2009

Abstract

In this paper a global and local optimization method is presented. This method is based on the integration of the heuristic Simulated Annealing, Tabu Search, Genetic Algorithms and Backpropagation. The performance of the method is investigated in the optimization of Multi-layer Perceptron artificial neural network architecture and weights. The heuristics perform the search in a constructive way and based on the pruning of irrelevant connections among the network nodes. Experiments demonstrated that the method can also be used for relevant feature selection. Experiments are performed with four classification and one prediction datasets.

References

  1. Anderson, E. (1953). The irises of the gasp peninsula. Bulletin of the American Iris Society, 59:2-5.
  2. Blake, C. L. and Merz, C. J. (1998). Uci repository of machine learning databases. Technical report, University of California, Irvine, CA.
  3. de Souza, J., Neto, B., dos Santos, F., C. de Melo, M. S., and Ludermir, T. (1999). Polypyrrole based aroma sensor. Synthetic Metals, 102:1296-1299.
  4. Mackey, M. C. and Glass, L. (1977). Oscillation and chaos in physic. control systems. Science, 197:287-289.
  5. Montgomery, D. C. (2001). Design and Analysis of Experiments (Fifth Edition). John Wiley and Sons.
  6. Quinlan, J. (1987). Simplifying decision trees. Int. Journal of Man-Machine Studies, 27:221234.
  7. Zanchettin, C. and Ludermir, T. B. (2006). A methodology to train and improve artificial neural networks weights and connections. In Int. Joint Conf. on Neural Networks, pages 5267-5274. IEEE Computer Society. 1The authors would like to thank CNPq (Brazilian re-
Download


Paper Citation


in Harvard Style

Zanchettin C. and B. Ludermir T. (2009). HYBRID OPTIMIZATION TECHNIQUE FOR ARTIFICIAL NEURAL NETWORKS DESIGN . In Proceedings of the 11th International Conference on Enterprise Information Systems - Volume 2: ICEIS, ISBN 978-989-8111-85-2, pages 242-247. DOI: 10.5220/0002012102420247


in Bibtex Style

@conference{iceis09,
author={Cleber Zanchettin and Teresa B. Ludermir},
title={HYBRID OPTIMIZATION TECHNIQUE FOR ARTIFICIAL NEURAL NETWORKS DESIGN},
booktitle={Proceedings of the 11th International Conference on Enterprise Information Systems - Volume 2: ICEIS,},
year={2009},
pages={242-247},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002012102420247},
isbn={978-989-8111-85-2},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 11th International Conference on Enterprise Information Systems - Volume 2: ICEIS,
TI - HYBRID OPTIMIZATION TECHNIQUE FOR ARTIFICIAL NEURAL NETWORKS DESIGN
SN - 978-989-8111-85-2
AU - Zanchettin C.
AU - B. Ludermir T.
PY - 2009
SP - 242
EP - 247
DO - 10.5220/0002012102420247