Abstract
We investigate the capability of multilayer perceptrons using specific risk functionals attaining the minimum probability of error (optimal performance) achievable by the class of mappings implemented by the multilayer perceptron (MLP). For that purpose we have carried out a large set of experiments using different risk functionals and datasets. The experiments were rigorously controlled so that any performance difference could only be attributed to the different risk functional being used. Statistical analysis was also conducted in a careful way. From the several conclusions that can be drawn from our experimental results it is worth to emphasize that a risk functional based on a specially tuned exponentially weighted distance attained the best performance in a large variety of datasets. As to the issue of attaining the minimum probability of error we also carried out classification experiments using non-MLP classifiers that implement complex mappings and are known to provide the best results until this date. These experiments have provided evidence that at least in many cases, by using an adequate risk functional, it will be possible to reach the optimal performance.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Silva, L.M., Alexandre, L.A., Marques de Sá, J.: Data classification with multilayer perceptrons using a generalized error function. Neural Networks (2008) doi:10.1016/j.neunet.2008.04,004
Erdogmus, D., Príncipe, J.C.: Generalized information potential criterion for adaptive system training. IEEE Transactions on Neural Networks 13(5), 1035–1044 (2002)
Erdogmus, D., Príncipe, J.C.: An Error-Entropy Minimization Algorithm for Supervised Training of Nonlinear Adaptive Systems. IEEE Transactions on Signal Processing 50(7), 1780–1786 (2002)
Santos, J.M., Alexandre, L.A., Marques de Sá, J.: The Error Entropy Minimization Algorithm for Neural Network Classification. In: Int. Conf. on Recent Advances in Soft Computing, Nottingham, United Kingdom (2004)
Duda, R., Hart, P., Stork, D.: Pattern Classification. Wiley-Interscience, Chichester (2001)
Newman, D., Hettich, S., Blake, C., Merz, C.: UCI repository of machine learning databases (1998)
Baum, E.B., Haussler, D.: What size net gives valid generalization? Neural Computation 1(1), 151–160 (1990)
Santos, J.M., Marques de Sá, J., Alexandre, L.A.: Neural Networks Trained with the EEM Algorithm: Tuning the Smoothing Parameter. WSEAS Transactions on Systems 4(4), 295–300 (2005)
Marques de Sá, J.: Using SPSS, STATISTICA, MATLAB and R, 2nd edn. Springer, Heidelberg (2007)
Embrechts, M.J., Szymanski, B., Sternickel, M.: Introduction to Scientific Data Mining: Direct Kernel Methods and Applications. In: Computationally Intelligent Hybrid Systems, ch. 10, pp. 317–363. Wiley Interscience, Chichester (2004)
Bennett, K.P., Embrechts, M.J.: An Optimization Perspective on Kernel Partial Least Squares Regression. In: Advances in Learning Theory: Methods, Models and Applications. NATO Science Series, Series III: Computer and System Sciences, vol. 190, pp. 227–249. IOS Press, Amsterdam (2003)
Suykens, J.A.K., Vandewalle, J.: Least squares support vector machine classifiers. Neural Processing Letters 9(3), 293–300 (1999)
Suykens, J.A.K., Gestel, T.V., Brabanter, J.D., Moor, B.D., Vandewalle, J.: Least Squares Support Vector Machines. World Scientific Pub. Co., Singapore (2003)
Keerthi, S.S., Shevade, S.K.: SMO algorithm for least squares SVM formulations. Neural Computation 15, 487–507 (2003)
Cristianini, N., Shawe-Taylor, J.: Support Vector Machines and Other Kernel-Based Learning Methods. Cambridge University Press, Cambridge (2000)
Vapnik, V.: Statistical Learning Theory. Wiley-Interscience, Chichester (1998)
Schölkopf, B., Smola, A.J.: Learning with Kernels. MIT Press, Cambridge (2002)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Silva, L.M., Embrechts, M., Santos, J.M., de Sá, J.M. (2008). The Influence of the Risk Functional in Data Classification with MLPs. In: Kůrková, V., Neruda, R., Koutník, J. (eds) Artificial Neural Networks - ICANN 2008. ICANN 2008. Lecture Notes in Computer Science, vol 5163. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-87536-9_20
Download citation
DOI: https://doi.org/10.1007/978-3-540-87536-9_20
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-87535-2
Online ISBN: 978-3-540-87536-9
eBook Packages: Computer ScienceComputer Science (R0)