 |
|
|
|
| HAL : hal-00270806, version 1 |
| arXiv : 0804.1026 |
| Fiche détaillée |  Récupérer au format |
|
|
|
|
| Testing for Homogeneity with Kernel Fisher Discriminant Analysis |
|
|
| Zaid Harchaoui 1Francis Bach 2 |
|
|
| (07/04/2008) |
|
|
|
| We propose to investigate test statistics for testing homogeneity in reproducing kernel Hilbert spaces. Asymptotic null distributions under null hypothesis are derived, and consistency against fixed and local alternatives is assessed. Finally, experimental evidence of the performance of the proposed approach on both artificial data and a speaker verification task is provided. |
|
|
|
|
|
|
|
|
|
|
|
| 1 : | Laboratoire traitement et communication de l'information (LTCI) |
|
|
|
CNRS : UMR5141 – Institut Télécom – Télécom ParisTech |
|
|
| 2 : | WILLOW (INRIA Rocquencourt) |
|
|
|
INRIA – Ecole Normale Supérieure de Paris - ENS Paris – Ecole Nationale des Ponts et Chaussées – CNRS : UMR8548 |
|
|
|
|
|
|
|
|
|
| Domaine | : | Statistiques/Apprentissage Machine |
|
|
| statistical hypothesis testing – reproducing kernel Hilbert space – covariance operator |
|
|
|
| Liste des fichiers attachés à ce document : | ||||||||||
|
||||||||||
|
|
|
| hal-00270806, version 1 | |
| http://hal.archives-ouvertes.fr/hal-00270806/fr/ | |
| oai:hal.archives-ouvertes.fr:hal-00270806_v1 | |
| Contributeur : Harchaoui Zaid | |
| Soumis le : Lundi 7 Avril 2008, 15:36:35 | |
| Dernière modification le : Lundi 7 Avril 2008, 16:17:35 | |
