Density Networks and their application to Protein Modelling

David J C MacKay

I define a latent variable model in the form of a neural network for which only target outputs are specified; the inputs are unspecified. Although the inputs are missing, it is still possible to train this model by placing a simple probability distribution on the unknown inputs and maximizing the probability of the data given the parameters. The model can then discover for itself a description of the data in terms of an underlying latent variable space of lower dimensionality. I present preliminary results of the application of these models to protein data.

postscript. (130K) | pdf.

@INPROCEEDINGS{MacKay95:density_nets,
 KEY ="",
 AUTHOR ="D. J. C. MacKay",
 TITLE ="Density Networks and their Application to Protein Modelling", 
 BOOKTITLE ="Maximum Entropy and {B}ayesian Methods, 
			{C}ambridge 1994",
 EDITOR 	="J. Skilling and S. Sibisi",
 PUBLISHER	="Kluwer",
 ADDRESS	="Dordrecht",
 YEAR ="1996",
 PAGES		="259-268",
 ANNOTE ="Date submitted: ; Date accepted: ; Collaborating institutes:
		 MRC Laboratory of Molecular Biology, Cambridge. MRAO 1837"}

David MacKay's: home page, publications.