article · QUT ePrints (Queensland University of Technology) · 2000

Large margin vector quantization

Cited by 0Abstract and details only

Abstract

In this paper we describe the Large Margin Vector Quantization algorithm (LMVQ), which uses gradient ascent to maximise the margin of a radial basis function classifier. We present a derivation of the algorithm, which proceeds from an estimate of the class-conditional probability densities. We show that the key behaviour of Kohonen's well-known LVQ2 and LVQ3 algorithms emerge as natural consequences of our formulation. We compare the performance of LMVQ with that of Kohonen's LVQ algorithms on an artificial classification problem and several well known benchmark classification tasks. We find that the classifiers produced by LMVQ attain a level of accuracy that compares well with those obtained via LVQ1, LVQ2 and LVQ3, with reduced storage complexity. We indicate future directions of enquiry based on the large margin approach to Learning Vector Quantization.

Cite this paper

APA 7

Buckingham, L., & Geva, S. (2000). Large margin vector quantization. QUT ePrints (Queensland University of Technology).

MLA 9

Buckingham, Lawrence, and Shlomo Geva. “Large margin vector quantization.” QUT ePrints (Queensland University of Technology), 2000.

Chicago

Buckingham, Lawrence, and Shlomo Geva. 2000. “Large margin vector quantization.” QUT ePrints (Queensland University of Technology).

Harvard

Buckingham, L. and Geva, S. (2000) ‘Large margin vector quantization’, QUT ePrints (Queensland University of Technology).

BibTeX

@article{buckingham2000large,
  title = {{Large margin vector quantization}},
  author = {Buckingham, Lawrence and Geva, Shlomo},
  journal = {QUT ePrints (Queensland University of Technology)},
  year = {2000}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.