Testing for Homogeneity with Kernel Fisher Discriminant Analysis
Zaid Harchaoui, Francis Bach and Eric Moulines
In: NIPS 2007, December 2007, Vancouver, Canada.
We propose to investigate test statistics for testing homogeneity based on kernel
Fisher discriminant analysis. Asymptotic null distribution under null hypothesis
is derived, and consistency against fixed alternatives is assessed. Finally, experimental
evidence of the performance of the proposed approach on artificial data
and on a speaker verification task is provided.