PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Auto-setting the SVM hyper-parameters using regularization paths
Gaëlle Loosli and Stéphane Canu
In: A Workshop for Women in Machine Learning, 04 october 2006, San Diego, USA.

Abstract

The main goal of this presentation is the auto-setting of the regularization parameter for SVM. To do so we are using the new Simple-nu-SVM solver which we are introducing, regularization path for SVM and on optimized leave-one-out validation method.

EPrint Type:Conference or Workshop Item (Poster)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:2952
Deposited By:Gaëlle Loosli
Deposited On:19 January 2007