Sequential Support Vector Classifiers and Regression

Support Vector Machines(SVMs) map the inputtraining data into a high dimensional feature spaceand finds a maximal margin hyperplane separatingthe data in that feature space. Extensions of thisapproach account for non-separable or noisy trainingdata (soft classifiers) as well as support vector basedregression. The optimal hyperplane is usually foundby solving a quadratic programming problem whichis usually quite complex, time consuming and proneto numerical instabilities. In this work,...