Dynamically Adapting Kernels in Support Vector Machines

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 and its value is usually found by means of a validation set. We present an algorithm which can automatically perform model selection with little additional computational cost and with no need of a validation set. In this procedure model selection and learning are not separate, but kernels are dynamically adjusted during the learning process to nd the kernel parameter which provides the best possible upper bound on the generalisation error. Theoretical results motivating the approach and experimental results con rming its validity are presented.

• Type: Conference
• Year: 1998
• Where: NIPS
• Authors: Nello Cristianini, Colin Campbell, John Shawe-Taylor
• Keywords: Model Selection, NIPS 1998, NIPS 2007, Procedure Model Selection, Validation Set