A signal theory approach to support vector classification: The sinc kernel

Authors:
James D. B. Nelson;Robert I. Damper;Steve R. Gunn;Baofeng Guo
Affiliations:
Information: Signals, Images, Systems Research Group, School of Electronics and Computer Science, University of Southampton, Southampton SO17 1BJ, UK;Information: Signals, Images, Systems Research Group, School of Electronics and Computer Science, University of Southampton, Southampton SO17 1BJ, UK;Information: Signals, Images, Systems Research Group, School of Electronics and Computer Science, University of Southampton, Southampton SO17 1BJ, UK;Information: Signals, Images, Systems Research Group, School of Electronics and Computer Science, University of Southampton, Southampton SO17 1BJ, UK
Venue:
Neural Networks
Year:
2009

Citing 8
Cited 0

Discrete-time signal processing

Discrete-time signal processing
The connection between regularization operators and support vector kernels

Neural Networks
An equivalence between sparse approximation and support vector machines

Neural Computation
Fast Transforms: Algorithms, Analyses, Applications

Fast Transforms: Algorithms, Analyses, Applications
Choosing Multiple Parameters for Support Vector Machines

Machine Learning
The subspace information criterion for infinite dimensional hypothesis spaces

The Journal of Machine Learning Research
Learning the Kernel Matrix with Semidefinite Programming

The Journal of Machine Learning Research
Semi-Supervised Learning on Riemannian Manifolds

Machine Learning

Quantified Score

Hi-index	0.00

Visualization

Abstract

Fourier-based regularisation is considered for the support vector machine classification problem over absolutely integrable loss functions. By invoking the modest assumption that the decision function belongs to a Paley-Wiener space, it is shown that the classification problem can be developed in the context of signal theory. Furthermore, by employing the Paley-Wiener reproducing kernel, namely the sinc function, it is shown that a principled and finite kernel hyper-parameter search space can be discerned, a priori. Subsequent simulations performed on a commonly-available hyperspectral image data set reveal that the approach yields results that surpass state-of-the-art benchmarks.