Offering a fundamental basis in kernel-based learning theory, this book covers both statistical and algebraic principles. It provides over 30 major theorems for kernel-based supervised and unsupervised learning models. The first of the theorems establishes a condition, arguably necessary and sufficient, for the kernelization of learning models. In addition, several other theorems are devoted to proving mathematical equivalence between seemingly unrelated models. With over 25 closed-form and iterative algorithms, the book provides a step-by-step guide to algorithmic procedures and analysing which factors to consider in tackling a given problem, enabling readers to improve specifically designed learning algorithms, build models for new applications and develop efficient techniques suitable for green machine learning technologies. Numerous real-world examples and over 200 problems, several of which are Matlab-based simulation exercises, make this an essential resource for graduate students and professionals in computer science, electrical and biomedical engineering. Solutions to problems are provided online for instructors.

A comprehensive introduction to Support Vector Machines and related kernel methods. In the 1990s, a new type of learning algorithm was developed, based on results from statistical learning theory: the Support Vector Machine (SVM). This gave rise to a new class of theoretically elegant learning machines that use a central concept of SVMs—-kernels—for a number of learning tasks. Kernel machines provide a modular framework that can be adapted to different tasks and domains by the choice of the kernel function and the base algorithm. They are replacing neural networks in a variety of fields, including engineering, information retrieval, and bioinformatics. Learning with Kernels provides an introduction to SVMs and related kernel methods. Although the book begins with the basics, it also includes the latest research. It provides all of the concepts necessary to enable a reader equipped with some basic mathematical knowledge to enter the world of machine learning using theoretically well-founded yet easy-to-use kernel algorithms and to understand and apply the powerful algorithms that have been developed over the last few years.

Publisher Description

A detailed overview of current research in kernel methods and their application to computational biology.

Support vector machines (SVMs) represent a breakthrough in the theory of learning systems. It is a new generation of learning algorithms based on recent advances in statistical learning theory. Designed for the undergraduate students of computer science and engineering, this book provides a comprehensive introduction to the state-of-the-art algorithm and techniques in this field. It covers most of the well known algorithms supplemented with code and data. One Class, Multiclass and hierarchical SVMs are included which will help the students to solve any pattern classification problems with ease and that too in Excel. KEY FEATURES  Extensive coverage of Lagrangian duality and iterative methods for optimization  Separate chapters on kernel based spectral clustering, text mining, and other applications in computational linguistics and speech processing  A chapter on latest sequential minimization algorithms and its modifications to do online learning  Step-by-step method of solving the SVM based classification problem in Excel.  Kernel versions of PCA, CCA and ICA The CD accompanying the book includes animations on solving SVM training problem in Microsoft EXCEL and by using SVMLight software . In addition, Matlab codes are given for all the formulations of SVM along with the data sets mentioned in the exercise section of each chapter.

Few developments have influenced the field of computer vision in the last decade more than the introduction of statistical machine learning techniques. Particularly kernel-based classifiers, such as the support vector machine, have become indispensable tools, providing a unified framework for solving a wide range of image-related prediction tasks, including face recognition, object detection and action classification. By emphasizing the geometric intuition that all kernel methods rely on, Kernel Methods in Computer Vision provides an introduction to kernel-based machine learning techniques accessible to a wide audience including students, researchers and practitioners alike, without sacrificing mathematical correctness. It covers not only support vector machines but also less known techniques for kernel-based regression, outlier detection, clustering and dimensionality reduction. Additionally, it offers an outlook on recent developments in kernel methods that have not yet made it into the regular textbooks: structured prediction, dependency estimation and learning of the kernel function. Each topic is illustrated with examples of successful application in the computer vision literature, making Kernel Methods in Computer Vision a useful guide not only for those wanting to understand the working principles of kernel methods, but also for anyone wanting to apply them to real-life problems.

A graduate textbook that provides a unified treatment of machine learning methods and their applications in the environmental sciences.