Work

Kernelizing Vector Quantization Algorithms

Year: 2009
Type: conference-paper
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. 1
Source: HAL
Institution: Supélec
Language: English
Open Access status: green
OpenAlex ID: W151603394
FWCI: 0.454
Cites: 10
Cited by: 3
Related to: 10