Learning with kernels

Support vector machines, regularization, optimization, and beyond

Bernhard Scholkopf, Alexander J. Smola

644 pages, parution le 27/09/2002

Ajouter à une liste

Indisponible

Résumé

"Interesting and original. Learning with Kernels will make a fine textbook on this subject."
-Grace Wahba, Bascom Professor of Statistics, University of Wisconsin-Madison"This splendid book fills the need for a comprehensive treatment of kernel methods and support vector machines. It collects results, theorems, and discussions from disparate sources into one very accessible exposition. I am particularly impressed that the authors have included problem sets at the end of each chapter; such problems are not easy to construct, but add significantly to the value of the book for the student audience."
- Chris J. C. Burges, Microsoft ResearchContents

1 An Tutorial Introduction

I Concepts and Tools

2 Kernels
3 Risk and Loss Functions
4 Regularization
5 Elements of Statistical Learning Theory
6 Optimization

II Support Vector Machines

7 Pattern Recognition
8 Single-Class Problems: Quantile Estimation and Novelty Detection
9 Regression Estimation
10 Implementation
11 Incorporating Invariances
12 Learning Theory Revisited

III Kernel Methods

13 Designing Kernels
14 Kernel Feature Extraction
15 Kernel Fisher Discriminant
16 Bayesian Kernel Methods
17 Regularized Principal Manifolds
18 Pre-Images and Reduced Set Methods

Caractéristiques techniques

	PAPIER
Éditeur(s)	The MIT Press
Auteur(s)	Bernhard Scholkopf, Alexander J. Smola
Parution	27/09/2002
Nb. de pages	644
Format	21 x 26
Couverture	Relié
Poids	1505g
Intérieur	Noir et Blanc
EAN13	9780262194754
ISBN13	978-0-262-19475-4