Boosting ─ Foundations and Algorithms

Boosting is an approach to machine learning based on the idea of creating a highlyaccurate predictor by combining many weak and inaccurate "rules of thumb." A remarkablyrich theory has evolved around boosting, with connections to a range of topics, includingstatistics, game theory, convex optimization, and information geometry. Boosting algorithms havealso enjoyed practical success in such fields as biology, vision, and speech processing. At varioustimes in its history, boosting has been perceived as mysterious, controversial, evenparadoxical.

This book, written by the inventors of the method, brings together,organizes, simplifies, and substantially extends two decades of research on boosting, presentingboth theory and applications in a way that is accessible to readers from diverse backgrounds whilealso providing an authoritative reference for advanced researchers. With its introductory treatmentof all material and its inclusion of exercises in every chapter, the book is appropriate for courseuse as well. The book begins with a general introduction to machine learning algorithms and theiranalysis; then explores the core theory of boosting, especially its ability to generalize; examinessome of the myriad other theoretical viewpoints that help to explain and understand boosting;provides practical extensions of boosting for more complex learning problems; and finally presents anumber of advanced theoretical topics. Numerous applications and practical illustrations are offeredthroughout.

Robert E. Schapire is Professor of Computer Science at Princeton University. Yoav Freund is Professor of Computer Science at the University of California, San Diego. For their work on boosting, Freund and Schapire received both the Godel Prize in 2003 and the Kanellakis Theory and Practice Award in 2004.