Kernelizing Vector Quantization Algorithms

Abstract : The kernel trick is a well known approach allowing to implicitly cast a linear method into a nonlinear one by replacing any dot product by a kernel function. However few vector quantization algorithms have been kernelized. Indeed, they usually imply to compute linear transformations (e.g. moving prototypes), what is not easily kernelizable. This paper introduces the Kernel-based Vector Quantization (KVQ) method which allows working in an approximation of the feature space, and thus kernelizing any Vector Quantization (VQ) algorithm.
Complete list of metadatas

https://hal-supelec.archives-ouvertes.fr/hal-00429892
Contributor : Sébastien van Luchene <>
Submitted on : Friday, December 4, 2009 - 12:00:29 PM
Last modification on : Wednesday, February 13, 2019 - 5:20:08 PM
Long-term archiving on : Thursday, June 17, 2010 - 6:02:01 PM

File

es2009-49.pdf
Publisher files allowed on an open archive

Identifiers

  • HAL Id : hal-00429892, version 1

Collections

Citation

Matthieu Geist, Olivier Pietquin, Gabriel Fricout. Kernelizing Vector Quantization Algorithms. ESANN'2009, Apr 2009, Bruges, Belgium. pp.541-546. ⟨hal-00429892⟩

Share

Metrics

Record views

119

Files downloads

96