Title: Bayesian Kernel Classification

Abstract: This is an on-going work about Bayesian kernel
classification. Unlike most kernel regression/classification models in
literature, our model allow kernel functions to have different shape
parameters. Those shape parameters give us great flexibility to model
different types of data. The approach uses alpha stable process priors
to promote sparsity in both the kernel functions and predictor
variables (or features); these priors act as regularizers for the
likelihood function that reward good selected features.  In large p
small n problems, we keep the kernel centers within the observed data
points, however, we could have more than one kernel centered at the
same location.