Forming Neural Networks Design through Evolution

Eva Volná

2007

Abstract

Neuroevolution techniques have been successful in many sequential decision tasks such as robot control and game playing. This paper aims at evolution in artificial neural networks (e.g. neuroevolution). Here is presented a neuroevolution system evolving populations of neurons that are combined to form the fully connected multilayer feedforward network with fixed architecture. In this paper, the transfer function has been shown to be an important part of architecture of the artificial neural network and have significant impact on an artificial neural network’s performance. In order to test the efficiency of described method, we applied it to the alphabet coding problem.

References

  1. Liu, Y. and Yao, X. “Evolutionary design of artificial neural networks with different nodes”. In Proc. 1996 IEEE Int. Conf. Evolutionary Computation (ICEC'96), Nagoya, Japan, pp. 670-675.
  2. Fogel, D. B., Evolutionary Computation: Toward a New Philosophy of Machine Intelligence. New York: IEEE Press, 1995.
  3. Bäck, T. Hammel, U.and Schwefel, H.-P.“Evolutionary computation: Comments on the history and current state,” IEEE Trans. Evolutionary Computation, vol. 1, pp. 3-17, Apr. 1997.
  4. Mani, G. “Learning by gradient descent in function space”. In Proc. IEEE Int. Conf. System, Man, and Cybernetics, Los Angeles, CA, 1990, pp. 242-247.
  5. Lovell D. R. and Tsoi, A. C. The Performance of the Neocognitron with Various S-Cell and C-Cell Transfer Functions, Intell. Machines Lab., Dep. Elect. Eng., Univ. Queensland, Tech. Rep., Apr. 1992.
  6. Stork, D. G. Walker, S. Burns, M. and Jackson, B. “Preadaptation in neural circuits”. In Proc. Int. Joint Conf. Neural Networks, vol. I, Washington, DC, 1990, pp. 202-205.
  7. White D. and Ligomenides, P. “GANNet: A genetic algorithm for optimizing topology and weights in neural network design”. In Proc. Int. Workshop Artificial Neural Networks (IWANN'93), Lecture Notes in Computer Science, vol. 686. Berlin, Germany: SpringerVerlag, 1993, pp. 322-327.
  8. Hwang, M. W. Choi, J. Y. and Park, J. “Evolutionary projection neural networks”. In Proc. 1997 IEEE Int. Conf. Evolutionary Computation, ICEC'97, pp. 667-671.
  9. Sebald, A. V. and Chellapilla, K. “On making problems evolutionarily friendly, part I: Evolving the most convenient representations”. In Porto, V. W., Saravanan, N., Waagen, D. and Eiben, A. E. (Eds.): Evolutionary Programming VII: Proc. 7th Annu Conf. Evolutionary Programming, vol. 1447 of Lecture Notes in Computer Science, Berlin, Germany: Springer-Verlag, 1998, pp. 271-280.
  10. Volná E. “Evolution design of an artificial neural network with fixed topology”, In R. Matoušek, P. Ošmera (eds.): Proceedings of the 12th International Conference on Soft Computing, Mendel'06, Brno, Czech Republic, 2006, pp. 1-6.
  11. Lawrence, D. Handbook of genetic algorithms. Van Nostrand Reinhold, New York 1991.
  12. Volná, E. „Using Neural network in cryptography“. In P. Sincák, J. Vašcák, V. Kvasnicka, R. Mesiar (eds.): The State of the Art in Computational Intelligence. Physica-Verlag Heidelberg 2000. pp.262-267.
  13. Garfinger, S. PGP: Pretty Good Privanci. Computer Press, Praha 1998.
  14. Volná, E. „Learning algorithm which learns both architectures and weights of feedforward neural networks“. Neural Network World. Int. Journal on Neural & Mass-Parallel Comp. and Inf. Systems. 8 (6): 653-664, 1998.
Download


Paper Citation


in Harvard Style

Volná E. (2007). Forming Neural Networks Design through Evolution . In Proceedings of the 3rd International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: ANNIIP, (ICINCO 2007) ISBN 978-972-8865-86-3, pages 13-20. DOI: 10.5220/0001624700130020


in Bibtex Style

@conference{anniip07,
author={Eva Volná},
title={Forming Neural Networks Design through Evolution},
booktitle={Proceedings of the 3rd International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: ANNIIP, (ICINCO 2007)},
year={2007},
pages={13-20},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001624700130020},
isbn={978-972-8865-86-3},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 3rd International Workshop on Artificial Neural Networks and Intelligent Information Processing - Volume 1: ANNIIP, (ICINCO 2007)
TI - Forming Neural Networks Design through Evolution
SN - 978-972-8865-86-3
AU - Volná E.
PY - 2007
SP - 13
EP - 20
DO - 10.5220/0001624700130020