Includes bibliographical references and index.

References: p. [361]-387.

Support vector machine solvers (Léon Bottou, Chih-Jen Lin) -- Training a support vector machine in the primal (Olivier Chapelle) -- Fast kernel learning with sparse inverted index (Patrick Haffner, Stephan Kanthak) -- Large-scale learning with string kernels (Sören Sonnenburg, Gunnar Rätsch, Konrad Rieck) -- Large-scale parallel SVM implementation (Igor Durdanovic, Eric Cosatto, Hans-Peter Graf) -- A distributed sequential solver for large-scale SVMs (Elad Yom-Tov) -- Newton methods for fast semisupervised linear SVMs (Vikas Sindhwani, S. Sathiya Keerthi) -- The improved fast Gauss transform with applications to machine learning (Vikas Chandrakant Raykar, Ramani Duraiswami) -- Approximation methods for gaussian process regression (Joaquin Quiñonero-Candela, Carl Edward Rasmussen) -- Brisk kernel independent component analysis (Stefanie Jegelka, Arthur Gretton) -- Building SVMs with reduced classifer complexity (S. Sathiya Keerthi, Olivier Chapelle, Dennis DeCoste) -- Trading convexity for scalability (Ronan Collobert, Fabian Sinz, Jasson Weston, Léon Bottou) -- Training invariant SVMs using selective sampling (Gaëlle Loosli, Léon Bottou, Stéphane Canu) -- Scaling learning algorithms toward AI (Yoshua Bengio, Yann LeCun).