FUSION 2008 Tutorial Proposal : Fundamentals of the Class - Specific Method

The class-specific method (CSM) is an approach for constructing classifiers with class-dependent features that blends signal processing with classifcation theory. CSM is based on the mathematical identity called the probability density function (PDF) projection theorem that extends classical Bayesian theory. In contrast to conventional classifiers, feature extraction is an integral part of the theory. Because CSM does not need a common feature space, the dimensionality curse can be avoided. The tutorial covers fundamentals including generative and discriminative classiifers, the PDF projection theorem, classspecific modules and the chain rule. Many intuitive examples are providded. Some advanced examples are also covered.