CONF
Moerland-98.2a/IDIAP
A comparison of mixture models for density estimation
Moerland, Perry
EXTERNAL
https://publications.idiap.ch/attachments/papers/1999/moerland-density98.pdf
PUBLIC
https://publications.idiap.ch/index.php/publications/showcite/moerland-98.2
Related documents
Proceedings of the International Conference on Artificial Neural Networks (ICANN'99)
1
25-30
1999
London: IEE
(IDIAP-RR 98-14)
Gaussian mixture models (GMMs) are a popular tool for density estimation. However, these models are limited by the fact that they either impose strong constraints on the covariance matrices of the component densities or no constraints at all. This paper presents an experimental comparison of GMMs and the recently introduced mixtures of linear latent variable models. It is shown that the latter models are a more flexible alternative for GMMs and often lead to improved results.
REPORT
Moerland-98.2/IDIAP
Localized mixtures of experts
Moerland, Perry
Idiap-RR-14-1998
1998
IDIAP