Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지
DOI QR코드

DOI QR Code

NETLA Based Optimal Synthesis Method of Binary Neural Network for Pattern Recognition

  • Lee, Joon-Tark (School of Electrical, Electronic ＆ Computer Engineering, Dong-A University)
  • Published : 2004.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper describes an optimal synthesis method of binary neural network for pattern recognition. Our objective is to minimize the number of connections and the number of neurons in hidden layer by using a Newly Expanded and Truncated Learning Algorithm (NETLA) for the multilayered neural networks. The synthesis method in NETLA uses the Expanded Sum of Product (ESP) of the boolean expressions and is based on the multilayer perceptron. It has an ability to optimize a given binary neural network in the binary space without any iterative learning as the conventional Error Back Propagation (EBP) algorithm. Furthermore, NETLA can reduce the number of the required neurons in hidden layer and the number of connections. Therefore, this learning algorithm can speed up training for the pattern recognition problems. The superiority of NETLA to other learning algorithms is demonstrated by an practical application to the approximation problem of a circular region.

Keywords

References

  1. M. Shimada, T. Saito: A GA- Based Learning Algorithm for Binary Neural Networks, IEICE Trans. Fund., vol.E85-A, no.11, pp. 2544-2546, Nov. 2002
  2. N. S. Chaudhari, A. Tiwari: Extending ETL for multi-class output, International Conference on Neural Information Processing, pp.1777-1780, 2002
  3. A. W. Andersen, S. S.Christensen, T. M. Jorgensen: An Active Vision System for Robot Guidance using a Low Cost Neural Network Board. In European Robotics and Intelligent Systems Conference, pp480-488, 1994
  4. P. L. Bartlett, T. Downs: Using Random Weight to rain Multilayer Network of Hard-limiting Units. IEEE Trans. Neural Networks, pp.202-210, 1992 https://doi.org/10.1109/72.125861
  5. M. L. Brady, R. Rayhavan, J. Slawny: Back Propagation fails to separate where Perceptrons Succeed. IEEE Trans. Circuits Systems, pp. 665-674, 1989 https://doi.org/10.1109/31.31314
  6. J.H. Kim, S. K. Park, Han, H. Oh, M. S. Han: The Geometrical Learning of Binary Neural Network. IEEE Trans. Neural Networks. vol. 6.no. 1. pp 237-247, 1995 https://doi.org/10.1109/72.363432
  7. Z. Kohavi.: Switching and Finite Automata Theory 2nd Ed. McGraw-Hill, 1986
  8. S. Park, J. H. Kim, H. Chang: A Learning Algorithm for Discrete Multilayer Perceptron. in Proc. Int. Sysp. Circuit Systems. Singapore, 1991
  9. Donald L. Gray, Anthony N. Michel: A Training Algorithm for Binary Feedforward Neural Networks. IEEE Trans. Neural Networks),pp.176-194, 1992 https://doi.org/10.1109/72.125859
  10. J.T. Lee, S.K. Sung: Optimal Synthesis Method for Binary Neural Network using NETRA, Advances in Soft Computing AFSS 2002, pp.236-244, 2002