Kernel projection machine: a new tool for pattern recognition
This paper investigates the effect of Kernel Principal Component Analysis (KPCA) within the classification framework, essentially the regularization properties of this dimensionality reduction method. We draw our inspiration from the pioneering works of Scholkopf, Smola and Muller on this topic. The method they used is somewhat redundant froma regularization point of view and we propose a new algorithm called Kernel projection machine to avoid this redundancy, based on an analogy with the statistical framework of regression for a Gaussian white noise model. Preliminary experimental results show that this algorithm reaches the same performances as an SVM.