Abstract
C-Mantec is a recently introduced constructive algorithm that generates compact neural architectures with good generalization abilities. Nevertheless, it produces a discrete output value and this might be a drawback in certain situations. We propose in this work two approaches in order to obtain a continuous output network such as the output can be interpreted as the probability of a given pattern to belong to one of the output classes. The CC-Mantec approach utilizes a committee strategy and the results obtained both with the XOR Boolean function and with a set of benchmark functions shows the suitability of the approach, as an improvement over the standard C-Mantec algorithm is obtained in almost all cases.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Baum, E.B., Haussler, D.: What size net gives valid generalization? Neural Comput. 1(1), 151–160 (1989)
Frean, M.: A thermal perceptron learning rule. Neural Comput. 4(6), 946–957 (1992)
Gómez, I., Franco, L., Jerez, J.M.: Neural network architecture selection: can function complexity help? Neural Process. Lett. 30(2), 71–87 (2009)
Haykin, S.: Neural Networks: A Comprehensive Foundation. Macmillan, New York (1994)
do Carmo Nicoletti, M., Bertini Jr., J.R.: An empirical evaluation of constructive neural network algorithms in classification tasks. Int. J. Innov. Comput. Appl. 1(1), 2–13 (2007)
Perrone, M.P., Cooper, L.N.: When networks disagree: Ensemble methods for hybrid neural networks, pp. 126–142. Chapman and Hall (1993)
Subirats, J.L., Franco, L., Jerez, J.M.: C-mantec: A novel constructive neural network algorithm incorporating competition between neurons. Neural Netw. 26, 130–140 (2012)
Subirats, J.L., Jerez, J.M., Franco, L.: A new decomposition algorithm for threshold synthesis and generalization of boolean functions. IEEE Trans. on Circuits and Systems 55-I(10), 3188–3196 (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Subirats, J.L., Luque-Baena, R.M., Urda, D., Ortega-Zamorano, F., Jerez, J.M., Franco, L. (2013). Committee C-Mantec: A Probabilistic Constructive Neural Network. In: Rojas, I., Joya, G., Gabestany, J. (eds) Advances in Computational Intelligence. IWANN 2013. Lecture Notes in Computer Science, vol 7902. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-38679-4_33
Download citation
DOI: https://doi.org/10.1007/978-3-642-38679-4_33
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-38678-7
Online ISBN: 978-3-642-38679-4
eBook Packages: Computer ScienceComputer Science (R0)