Selecting relevant features for Support Vector Machine (SVM) classifiers is important for a variety of reasons such as generalization performance, computational efficiency, and ...
Practical experience has shown that in order to obtain the best possible performance, prior knowledge about invariances of a classification problem at hand ought to be incorporated...
The kernel-parameter is one of the few tunable parameters in Support Vector machines, controlling the complexity of the resulting hypothesis. Its choice amounts to model selection...
Nello Cristianini, Colin Campbell, John Shawe-Tayl...
We introduce a framework, which we call Divide-by-2 (DB2), for extending support vector machines (SVM) to multi-class problems. DB2 offers an alternative to the standard one-again...
We propose a novel approach that reduces cost-sensitive classification to one-sided regression. The approach stores the cost information in the regression labels and encodes the m...