Gaussian Mixture Modeling with Gaussian Process Latent Variable Models
PDF Web
Density modeling is notoriously difficult for high dimensional data. One approach to the problem is to search for a lower dimensional manifold which captures the main characteristics of the data. Recently, the Gaussian Process Latent Variable Model (GPLVM) has successfully been used to find low dimensional manifolds in a variety of complex data. The GPLVM consists of a set of points in a low dimensional latent space, and a stochastic map to the observed space. We show how it can be interpreted as a density model in the observed space. However, the GPLVM is not trained as a density model and therefore yields bad density estimates. We propose a new training strategy and obtain improved generalisation performance and better density estimates in comparative evaluations on several benchmark data sets.
| Author(s): | Nickisch, H. and Rasmussen, CE. |
| Links: | |
| Book Title: | Pattern Recognition |
| Journal: | Pattern Recognition: 32nd DAGM Symposium |
| Pages: | 271-282 |
| Year: | 2010 |
| Month: | September |
| Day: | 0 |
| Editors: | Goesele, M. , S. Roth, A. Kuijper, B. Schiele, K. Schindler |
| Publisher: | Springer |
| BibTeX Type: | Conference Paper (inproceedings) |
| Address: | Berlin, Germany |
| DOI: | 10.1007/978-3-642-15986-2_28 |
| Event Name: | 32nd Annual Symposium of the German Association for Pattern Recognition (DAGM 2010) |
| Event Place: | Darmstadt, Germany |
| Digital: | 0 |
| Electronic Archiving: | grant_archive |
| Institution: | Deutsche Arbeitsgemeinschaft für Mustererkennung |
| ISBN: | 978-3-642-15986-2 |
| Language: | en |
| Organization: | Max-Planck-Gesellschaft |
| School: | Biologische Kybernetik |
BibTeX
@inproceedings{6716,
title = {Gaussian Mixture Modeling with Gaussian Process Latent Variable Models},
journal = {Pattern Recognition: 32nd DAGM Symposium},
booktitle = {Pattern Recognition},
abstract = {Density modeling is notoriously difficult for high dimensional data. One approach to the problem is to search for a lower dimensional manifold which captures the main characteristics of the data. Recently, the Gaussian Process Latent Variable Model (GPLVM) has successfully been used to find low dimensional manifolds in a variety of complex data. The GPLVM consists of a set of points in a low dimensional latent space, and a stochastic map to the observed space. We show how it can be interpreted as a density model in the observed space. However, the GPLVM is not trained as a density model and therefore yields bad density estimates. We propose a new training strategy and obtain improved generalisation performance and better density estimates in comparative evaluations on several benchmark data sets.},
pages = {271-282},
editors = {Goesele, M. , S. Roth, A. Kuijper, B. Schiele, K. Schindler},
publisher = {Springer},
organization = {Max-Planck-Gesellschaft},
institution = {Deutsche Arbeitsgemeinschaft für Mustererkennung},
school = {Biologische Kybernetik},
address = {Berlin, Germany},
month = sep,
year = {2010},
author = {Nickisch, H. and Rasmussen, CE.},
doi = {10.1007/978-3-642-15986-2_28},
month_numeric = {9}
}