Optimal vector quantization in terms of Wasserstein distance

  • The optimal quantizer in memory-size constrained vector quantization induces a quantization error which is equal to a Wasserstein distortion. However, for the optimal (Shannon-)entropy constrained quantization error a proof for a similar identity is still missing. Relying on principal results of the optimal mass transportation theory, we will prove that the optimal quantization error is equal to a Wasserstein distance. Since we will state the quantization problem in a very general setting, our approach includes the R\'enyi-$\alpha$-entropy as a complexity constraint, which includes the special case of (Shannon-)entropy constrained $(\alpha = 1)$ and memory-size constrained $(\alpha = 0)$ quantization. Additionally, we will derive for certain distance functions codecell convexity for quantizers with a finite codebook. Using other methods, this regularity in codecell geometry has already been proved earlier by Gy\"{o}rgy and Linder.

Download full text files

Export metadata

Additional Services

Search Google Scholar
Metadaten
Author:Wolfgang Kreitmeier
URN:urn:nbn:de:bvb:739-opus-22502
Document Type:Preprint
Language:English
Year of Completion:2011
Date of Publication (online):2011/04/18
Publishing Institution:Universität Passau
Release Date:2011/04/18
Tag:R\'enyi-$\alpha$-entropy; Wasserstein distance; codecell convexity; optimal quantization error
GND Keyword:Maßtheorie; Transporttheorie; Quantisierung; Entropie
Note:
This is a preprint of an article accepted for publication in the Journal of Multivariate Analysis ISSN 0047-259X. The original publication is available at http://www.elsevier.com/. The digital object identifier (DOI) of the definitive article is 10.1016/j.jmva.2011.04.005.
Source:Journal of Multivariate Analysis. ISSN 0047-259X
Institutes:Fakultät für Informatik und Mathematik / Mitarbeiter Lehrstuhl/Einrichtung der Fakultät für Informatik und Mathematik
Dewey Decimal Classification:5 Naturwissenschaften und Mathematik / 51 Mathematik / 510 Mathematik
open_access (DINI-Set):open_access
Licence (German):License LogoStandardbedingung laut Einverständniserklärung