[BiO BB] [Fwd: LS-SVMs: book announcement and related LS-SVMlab software]

Johan Suykens Johan.Suykens at esat.kuleuven.ac.be
Thu Dec 5 05:32:31 EST 2002

We are glad to announce the publication of a new book


J.A.K. Suykens, T. Van Gestel, J. De Brabanter, B. De Moor, J. Vandewalle,
Least Squares Support Vector Machines,
World Scientific Pub. Co., Singapore, 2002


This book focuses on Least Squares Support Vector Machines (LS-SVMs)
which are reformulations to standard SVMs. LS-SVMs are closely related
to regularization networks and Gaussian processes but additionally
emphasize and exploit primal-dual interpretations from optimization theory.
The authors explain the natural links between LS-SVM classifiers and kernel
Fisher discriminant analysis. Bayesian inference of LS-SVM models is
discussed, together with methods for imposing sparseness and employing
robust statistics.

The framework is further extended towards unsupervised learning by
considering PCA analysis and its kernel version as a one-class modelling
problem. This leads to new primal-dual support vector machine formulations
for kernel PCA and kernel CCA analysis. Furthermore, LS-SVM formulations
are given for recurrent networks and control. In general, support vector
machines may pose heavy computational challenges for large data sets.
For this purpose, a method of fixed size LS-SVM is proposed where the
estimation is done in the primal space in relation to a Nyström sampling
with active selection of support vectors. The methods are illustrated
with several examples.

. Introduction
. Support vector machines
. Least squares support vector machines, links with Gaussian
  processes, regularization networks, and kernel FDA
. Bayesian inference for LS-SVM models
. Weighted versions and robust statistics
. Large scale problems: Nystrom sampling, reduced set methods,
  basis formation and Fixed size LS-SVM
. LS-SVM for unsupervised learning: support vector machines
  formulations for kernel PCA. Related methods of kernel CCA.
. LS-SVM for recurrent networks and control
. Illustrations and applications

Graduate students and researchers in neural networks; machine learning;
data-mining; signal processing; circuit, systems and control theory;
pattern recognition; and statistics.

Info: 308pp., Publication date: Nov. 2002,
ISBN 981-238-151-1

Order information: World Scientific


Least Squares - Support Vector Machines Matlab/C Toolbox


. Matlab LS-SVMlab1.4 - Linux and Windows Matlab/C code
. Basic and advanced versions
. Functional and object oriented interface

Tutorial User's Guide (100pp.):
. Examples and demos
. Matlab functions with help

Solving and handling:
. Classification, Regression
. Tuning, cross-validation, fast loo,
 receiver operating characteristic (ROC) curves
. Small and unbalanced data sets
. High dimensional input data
. Bayesian framework with three levels of inference
. Probabilistic interpretations, error bars
. hyperparameter selection, automatic relevance determination (ARD)
 input selection, model comparison
. Multi-class encoding/decoding
. Sparseness
. Robustness, robust weighting, robust cross-validation
. Time series prediction
. Fixed size LS-SVM, Nystrom method,
 kernel principal component analayis (kPCA), ridge regression
. Unsupervised learning
. Large scale problems

Related links, publications, presentations and book:

Contact: LS-SVMlab at esat.kuleuven.ac.be

GNU General Public License:
The LS-SVMlab software is made available for research purposes only
under the GNU General Public License. LS-SVMlab software may not be
used for commercial purposes without explicit written permission after
contacting LS-SVMlab at esat.kuleuven.ac.be .

[we apologize for receiving multiple copies of this message]

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bioinformatics.org/pipermail/bbb/attachments/20021205/a6601e06/attachment.html>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bioinformatics.org/pipermail/bbb/attachments/20021205/a6601e06/attachment-0001.html>

More information about the BBB mailing list