Abstract
In most treatments of the regression problem it is assumed that the distribution of target data can be described by a deterministic function of the inputs, together with additive Gaussian noise having constant variance. The use of maximum likelihood to train such models then corresponds to the minimization of a sum-of-squares error function. In many applications a more realistic model would allow the noise variance itself to depend on the input variables. However, the use of maximum likelihood for training such models would give highly biased results. In this paper we show how a Bayesian treatment can allow for an input-dependent variance while overcoming the bias of maximum likelihood.
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
Bishop, C. M. (1995). Neural Networks for Pattern Recognition. Oxford University Press.
MacKay, D. J. C. (1991). Bayesian Methods for Adaptive Models. Ph.D. thesis, California Institute of Technology.
MacKay, D. J. C. (1995). Probabilistic networks: new models and new methods. In F. Fogelman-Soulié and P. Gallinari (Eds.), Proceedings ICANN'95 International Conference on Artificial Neural Networks, pp. 331–337. Paris: EC2 & Cie.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bishop, C.M., Qazaz, C.S. (1996). Bayesian inference of noise levels in regression. In: von der Malsburg, C., von Seelen, W., Vorbrüggen, J.C., Sendhoff, B. (eds) Artificial Neural Networks — ICANN 96. ICANN 1996. Lecture Notes in Computer Science, vol 1112. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-61510-5_14
Download citation
DOI: https://doi.org/10.1007/3-540-61510-5_14
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61510-1
Online ISBN: 978-3-540-68684-2
eBook Packages: Springer Book Archive