Covariance Kernels from Bayesian Generative Models

We propose the framework of mutual information kernels for learning covariance kernels, as used in Support Vector machines and Gaussian process classifiers, from unlabeled task data using Bayesian techniques. We describe an implementation of this framework which uses variational Bayesian mixtures of factor analyzers in order to attack classification problems in high-dimensional spaces where labeled data is sparse, but unlabeled data is abundant.


Published in:
Proceedings of the 15th Annual Conference on Neural Information Processing Systems, 905-912
Presented at:
Neural Information Processing Systems 14, Vancouver, BC
Year:
2002
Keywords:
Laboratories:




 Record created 2010-12-01, last modified 2018-09-25

n/a:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)