Abstract
This Letter discusses the application of gradient-based methods to train a single layer perceptron subject to the constraint that the saturation degree of the sigmoid activation function (measured as its maximum slope in the sample space) is fixed to a given value. From a theoretical standpoint, we show that, if the training set is not linearly separable, the minimization of an L p error norm provides an approximation to the minimum error classifier, provided that the perceptron is highly saturated. Moreover, if data are linearly separable, the perceptron approximates the maximum margin classifier
Similar content being viewed by others
References
Auer, P., Cesa-Bianchi, N. and Gentile, C.: Adaptive and Self-Concdent On-Line Learning Algorithms, NeuroCOLT Technical Report NC-TR-00-083. Submitted for journal publication, 2001.
Bobrowsky, L. and Sklansky, J.: Linear Classicers by Window Training, IEEE Transactions on System Science and Cybernetics 25 (1995), 1–9.
Do-Tu, H. and Installe, M.: Learning Algorithms for Non-Parametric Solution to the Minimum Error Classification Problem, IEEE Transactions on Computers 27(7) (1978), 648–659.
Duda, R. O. and Hart, P. E.: Pattern Classification and Scene Analysis, Wiley-Interscience, 1973.
Navia-Vázquez, A., Pérez-Cruz, F., Artés-Rodríguez, A. and Figueiras-Vidal, A. R.: Weighted Least Squares Training of Support Vector Classicers Leading to Compact and Adaptive Schemes, IEEE Transactions on Neural Networks. Accepted for publication. To appear, 2001.
Raudys, S.: Evolution and Generalization of a Single Neurone. I. Single Layer Perceptron as Seven Statistical Classicers, Neural Networks 11 (1998a), 283–296.
Raudys, S.: Evolution and Generalization of a Single Neurone. II. Complexity of Statistical Classicers and Sample Size Considerations, Neural Networks 11 (1998b), 297–313.
Ruck, D. W., Rogers, S. K., Kabrinsky, M., Oxley, M. E. and Suter, B. W.: The Multilayer Perceptron as an Approximation to a Bayes Optimal Discriminant Function, IEEE Transactions on Neural Networks 1(4) (1990), 296–298.
Telfer, B. A. and Szu, H. H.: Energy Functions for Minimizing Misclassiccation Error with Minimum-Complexity Networks, Neural Networks 7 (1994), 809–818.
Vapnik, V.: Statistical Learning Theory. New York: Wiley, 1998.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cid-Sueiro, J., Sancho-Gómez, J.L. Saturated Perceptrons for Maximum Margin and Minimum Misclassification Error. Neural Processing Letters 14, 217–226 (2001). https://doi.org/10.1023/A:1012755431700
Issue Date:
DOI: https://doi.org/10.1023/A:1012755431700