Author
Listed:
- Yoshua Bengio
- Pascal Vincent
Abstract
The similarity between objects is a fundamental element of many learning algorithms. Most non-parametric methods take this similarity to be fixed, but much recent work has shown the advantages of learning it, in particular to exploit the local invariances in the data or to capture the possibly non-linear manifold on which most of the data lies. We propose a new non-parametric kernel density estimation method which captures the local structure of an underlying manifold through the leading eigenvectors of regularized local covariance matrices. Experiments in density estimation show significant improvements with respect to Parzen density estimators. The density estimators can also be used within Bayes classifiers, yielding classification rates similar to SVMs and much superior to the Parzen classifier. La similarité entre objets est un élément fondamental de plusieurs algorithmes d'apprentissage. La plupart des méthodes non paramétriques supposent cette similarité constante, mais des travaux récents ont montré les avantages de les apprendre, en particulier pour exploiter les invariances locales dans les données ou pour capturer la variété possiblement non linéaire sur laquelle reposent la plupart des données. Nous proposons une nouvelle méthode d'estimation de densité à noyau non paramétrique qui capture la structure locale d'une variété sous-jacente en utilisant les vecteurs propres principaux de matrices de covariance locales régularisées. Les expériences d'estimation de densité montrent une amélioration significative sur les estimateurs de densité de Parzen. Les estimateurs de densité peuvent aussi être utilisés à l'intérieur de classificateurs de Bayes, menant à des taux de classification similaires à ceux des SVMs, et très supérieurs au classificateur de Parzen.
Suggested Citation
Yoshua Bengio & Pascal Vincent, 2004.
"Manifold Parzen Windows,"
CIRANO Working Papers
2004s-30, CIRANO.
Handle:
RePEc:cir:cirwor:2004s-30
Download full text from publisher
Corrections
All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:cir:cirwor:2004s-30. See general information about how to correct material in RePEc.
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
We have no bibliographic references for this item. You can help adding them by using this form .
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Webmaster (email available below). General contact details of provider: https://edirc.repec.org/data/ciranca.html .
Please note that corrections may take a couple of weeks to filter through
the various RePEc services.