Bayesian Neural Networks and Density Networks

David J C MacKay

This paper reviews the Bayesian approach to learning in neural networks, then introduces a new adaptive model, the density network. This is a neural network for which target outputs are provided, but the inputs are unspecified. When a probability distribution is placed on the unknown inputs, a latent variable model is defined that is capable of discovering the underlying dimensionality of a data set. A Bayesian learning algorithm for these networks is derived and demonstrated.

postscript (Cambridge UK).

postscript (Canada mirror).

@ARTICLE{MacKay95:wonsda,
 AUTHOR         ="D. J. C.  MacKay",
 TITLE          ="Bayesian Neural Networks and Density Networks", 
 JOURNAL        ="Nuclear Instruments and Methods in Physics 
                  Research, Section A",
 volume = 354,
 number = 1,
 YEAR           =1995,
 PAGES="73-80",
 ANNOTE ="Date submitted: 1994; Date accepted: 1994; Collaborating institutes: none"}

David MacKay's: home page, publications. bibtex file.

Canadian mirrors: home page, publications. bibtex file.