We propose a local, generative model for similarity-based classification. The method is applicable to the case that only pairwise similarities between samples are available. The c...
Incorporation of prior knowledge into the learning process can significantly improve low-sample classification accuracy. We show how to introduce prior knowledge into linear supp...
One way of providing technological support for CoPs is to help participants to produce, structure and share information. As this information becomes more and more multimedia in nat...
Sparse methods for supervised learning aim at finding good linear predictors from as few variables as possible, i.e., with small cardinality of their supports. This combinatorial ...
We study the problem of uncertainty in the entries of the Kernel matrix, arising in SVM formulation. Using Chance Constraint Programming and a novel large deviation inequality we ...