Abstract
We present in this paper a new neural net called Bi-directional Multi Layer Perceptron (BMLP). Its structure is identical to the one of the classical Multi Layer Perceptron (MLP) but it differs in its neurons. Those neurons possess the bi-directional connections. The BMLP is capable to implement non linear transformations and their reverses for functional approximation problems.
Similar content being viewed by others
References
Rosenblatt, F.: The perceptron: a probabilistic model for information storage and organization in the brain, Psychological Review 65 (1958), 386–408.
Rumelhart, D. E., Hinton, G. E. and Williams, R. J.: Learning internal representations by error propagation, In: Parallel Distributed Processing: Exploration in theMicrostructure of Cognition, Vol. 1, Cambridge, MA: MIT Press, 1986a.
Kosko, B.: Adaptive bi-directional associative memory, App. Opt. 26(23) (1987), 4947–4960.
Kosko, B.: Bi-directional associative memory, IEEE Trans. Syst. Man. Cybern. 18(1) (1988), 49–60.
MacCulloch, W. S. and Pitts, W.: A logical calculus of the ideas immanent in nervous activity, Bulletin of Mathematical Biophysics 5 (1943), 115–133.
Rumelhart, D. E., Hinton, G. E. and Williams, R. J.: Learning representations by backpropagating error, Nature 323 (1986b), 533–536.
Pao, Y. H.: Adaptative Pattern Recognition and Neural Networks, Addison-Wesley Publishing Company Inc., 1989.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Jedra, M., El Ouardighi, A., Essaid, A. et al. A Bi-Directional Multilayer Perceptron. Neural Processing Letters 10, 89–95 (1999). https://doi.org/10.1023/A:1018777200501
Issue Date:
DOI: https://doi.org/10.1023/A:1018777200501