Abstract
In this paper, we address the problem of selecting the proper architecture of Bayesian neural networks. Specifically, we propose a variable architecture model where input-to-hidden connections and, therefore, hidden units are selected by using a variant of the Evolutionary Monte Carlo algorithm developed by (2000). To perform the Bayesian learning of parameters we propose a hybrid Markov chain Monte Carlo algorithm which includes an Evolutionary Monte Carlo step for selecting the architecture. Simulation results show the effectiveness of the proposed approach.
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
BISHOP, C.M. (1995): Neural Networks for Pattern Recognition, Oxford University Press.
GILKS, W.R., RICHARDSON, S. and SPIEGELHALTER, D.J. (1996): Markov chain Monte Carlo in practice, Chapman & Hall.
KIRKPATRICK S., GELATT Jr., C.D. and VECCHI, M.P. (1983): Optimization by simulated annealing. Science, vol. 220, 671–680.
LIANG, F. and WONG, W.H. (2000): Evolutionary Monte Carlo: Applications to Cp Model Sampling and Change Point Problem. Statistica Sinica, vol. 10, 317–342.
LIANG, F. and WONG, W.H. (2001): Real-Parameter Evolutionary Monte Carlo with Applications to Bayesian Mixture Models. Journal of the American Statistical Association, vol. 96, 653–666.
MACKAY, D.J.C. (1996): A Practical Bayesian Framework for Backpropagation Networks. Neural Computation, vol. 4, 448–472.
MITCHELL, M. (1996): An Introduction to Genetic Algorithms. MIT Press.
MÜLLER, P. and RIOS INSUA, D. (1998): Issues in Bayesian Analysis of Neural Network Models. Neural Computation vol. 10, 749–770.
NEAL, R.M. (1996): Bayesian Learning for Neural Networks. Lecture Notes in Statistics. Springer.
RIOS INSUA, D. and MÜLLER, P. (1998): Feedforward Neural Networks for Nonparametric Regression. In Dey D.K., Müller, P. and Sinha D. (Eds.) Practical Nonparametric and Semiparametric Bayesian Statistics. Springer, 181–194.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bozza, S., Mantovan, P., Schiavo, R.A. (2003). Evolutionary Model Selection in Bayesian Neural Networks. In: Schader, M., Gaul, W., Vichi, M. (eds) Between Data Science and Applied Data Analysis. Studies in Classification, Data Analysis, and Knowledge Organization. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-18991-3_43
Download citation
DOI: https://doi.org/10.1007/978-3-642-18991-3_43
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-40354-8
Online ISBN: 978-3-642-18991-3
eBook Packages: Springer Book Archive