Abstract
Among binary unit-based constructive algorithms, the Sequential Learning is particularly interesting for many reasons, the most significant one being its ability to treat real valued inputs without preprocessing. However, due to the construction process, the classical algorithms derived from the Perceptron cannot be used for learning each unit of the hidden layer. The BCP Max, recently proposed, appears as the first efficient heuristic algorithm to perform the particular neuron training in the Sequential Learning. But the BCP Max principles can be easily extended to the classical Perceptron derivatives usually used in constructive algorithms. In this paper, we show how to extend the Thermal Perceptron, the Pocket algorithm, the Ratchet and the simple Perceptron to train a neuron in the Sequential Learning. Finally all solutions are compared.
Similar content being viewed by others
References
S. Gallant, “Optimal linear discriminants”,in IEEE Proc. 8th Conf. Pattern Recognition (Paris), pp. 849–852, 1986, IEEE Computer Society Press, Washington, DC.
S. Gallant,“Perceptron-based learning algorithms”, IEEE Transactions on Neural Networks, Vol. 1, pp. 179–191, 1990.
M. Frean,“A thermal perceptron learning rule”, Neural Computation, Vol. 4, No. 6, pp. 946–957, 1992.
M. Marchand, M. Goleaand P. Rújan, “A convergence theorem for sequential learning in two-layer perceptron”, Europhysics Letters, Vol. 11, pp. 487–492, 1990.
H. Poulard and N. Hernandez, “Two efficientmultiple-output constructive algorithms”, Technical Report LAAS-CNRS, 1996. http://www.laas.fr/_poulard/papers/
H. Poulard and S. Labrèche, “A new algorithm for learning threshold unit”, Technical Report LAAS-CNRS, 1996. http://www.laas.fr/_poulard/papers/
C. Campbell and C.P. Vicente, “The Target Switch algorithm: aconstructive learning procedure for feedforward neural networks”, Neural Computation, Vol. 7, pp. 1245–1264, 1995.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Poulard, H., Hernandez, N. Training a neuron in sequential learning. Neural Processing Letters 5, 29–34 (1997). https://doi.org/10.1023/A:1009605823241
Issue Date:
DOI: https://doi.org/10.1023/A:1009605823241