The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Cet article présente un nouvel algorithme d'apprentissage compétitif pour la conception de quantificateurs vectoriels (VQ) à taux variable. L'algorithme, appelé algorithme d'apprentissage compétitif à taux variable (VRCL), conçoit un VQ présentant une distorsion moyenne minimale soumise à une contrainte de débit. Le VRCL effectue l'entraînement du vecteur de poids dans le domaine des ondelettes afin que le temps d'entraînement requis soit court. De plus, l'algorithme bénéficie de meilleures performances de distorsion de débit que celles des autres algorithmes de conception VQ et algorithmes d'apprentissage compétitifs existants. L'algorithme d'apprentissage est également plus insensible à la sélection des mots de passe initiaux par rapport aux algorithmes de conception existants. Par conséquent, l’algorithme VRCL peut constituer une alternative efficace aux algorithmes de conception VQ à débit variable existants pour les applications de compression de signal.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Wen-Jyi HWANG, Maw-Rong LEOU, Shih-Chiang LIAO, Chienmin OU, "A Novel Competitive Learning Technique for the Design of Variable-Rate Vector Quantizers with Reproduction Vector Training in the Wavelet Domain" in IEICE TRANSACTIONS on Information,
vol. E83-D, no. 9, pp. 1781-1789, September 2000, doi: .
Abstract: This paper presents a novel competitive learning algorithm for the design of variable-rate vector quantizers (VQs). The algorithm, termed variable-rate competitive learning (VRCL) algorithm, designs a VQ having minimum average distortion subject to a rate constraint. The VRCL performs the weight vector training in the wavelet domain so that required training time is short. In addition, the algorithm enjoys a better rate-distortion performance than that of other existing VQ design algorithms and competitive learning algorithms. The learning algorithm is also more insensitive to the selection of initial codewords as compared with existing design algorithms. Therefore, the VRCL algorithm can be an effective alternative to the existing variable-rate VQ design algorithms for the applications of signal compression.
URL: https://global.ieice.org/en_transactions/information/10.1587/e83-d_9_1781/_p
Copier
@ARTICLE{e83-d_9_1781,
author={Wen-Jyi HWANG, Maw-Rong LEOU, Shih-Chiang LIAO, Chienmin OU, },
journal={IEICE TRANSACTIONS on Information},
title={A Novel Competitive Learning Technique for the Design of Variable-Rate Vector Quantizers with Reproduction Vector Training in the Wavelet Domain},
year={2000},
volume={E83-D},
number={9},
pages={1781-1789},
abstract={This paper presents a novel competitive learning algorithm for the design of variable-rate vector quantizers (VQs). The algorithm, termed variable-rate competitive learning (VRCL) algorithm, designs a VQ having minimum average distortion subject to a rate constraint. The VRCL performs the weight vector training in the wavelet domain so that required training time is short. In addition, the algorithm enjoys a better rate-distortion performance than that of other existing VQ design algorithms and competitive learning algorithms. The learning algorithm is also more insensitive to the selection of initial codewords as compared with existing design algorithms. Therefore, the VRCL algorithm can be an effective alternative to the existing variable-rate VQ design algorithms for the applications of signal compression.},
keywords={},
doi={},
ISSN={},
month={September},}
Copier
TY - JOUR
TI - A Novel Competitive Learning Technique for the Design of Variable-Rate Vector Quantizers with Reproduction Vector Training in the Wavelet Domain
T2 - IEICE TRANSACTIONS on Information
SP - 1781
EP - 1789
AU - Wen-Jyi HWANG
AU - Maw-Rong LEOU
AU - Shih-Chiang LIAO
AU - Chienmin OU
PY - 2000
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E83-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2000
AB - This paper presents a novel competitive learning algorithm for the design of variable-rate vector quantizers (VQs). The algorithm, termed variable-rate competitive learning (VRCL) algorithm, designs a VQ having minimum average distortion subject to a rate constraint. The VRCL performs the weight vector training in the wavelet domain so that required training time is short. In addition, the algorithm enjoys a better rate-distortion performance than that of other existing VQ design algorithms and competitive learning algorithms. The learning algorithm is also more insensitive to the selection of initial codewords as compared with existing design algorithms. Therefore, the VRCL algorithm can be an effective alternative to the existing variable-rate VQ design algorithms for the applications of signal compression.
ER -