Testing for Homogeneity with Kernel Fisher Discriminant Analysis

Zaid Harchaoui (1) , Francis Bach (2) , Eric Moulines (1)

1 LTCI - Laboratoire Traitement et Communication de l'Information
2 WILLOW - Models of visual object recognition and scene understanding

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.

Mots clés: statistical hypothesis testing; reproducing kernel Hilbert space; covariance operator

Domaines: Autres [stat.ML]

Zaid Harchaoui, Francis Bach, Eric Moulines. Testing for Homogeneity with Kernel Fisher Discriminant Analysis. 2008. ⟨hal-00270806⟩