Kernel Machines
Abstract
This chapter addresses two fundamental questions: First, how can we perform regression for complex nonlinear maps or separate patterns that are nonlinearly separable? Second, how can we generalize to new examples even in presence of small training sets? Both these two fundamental issues are addressed by introducing the elegant framework of kernel machines. After having discussed the construction of feature spaces, the reader is invited to face the classical maximum margin problem that naturally leads to the introduction of kernel machines. Finally, they are framed in the regularization framework by following the common formalism of the reproducing kernel Hilbert space, as well the related approach based ...
Get Machine Learning now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.