Vector quantization with optimized grouping and parallel distributed processing

Y. Matsuyama

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Vector quantization with optimized grouping of elements is studied. The presented vector quantization allows optimal or suboptimal grouping of source data. Thus, the algorithms herein are called variable region vector quantization. The optimization yielding the data subgroups can also be interpreted as the connection weight adjustmen. The presented methods are still executable on conventional SISD computers. However, the adaptation of the variable region vector quantization to SIMD and MIMD computation via PDP (Parallel Distributed Processing) approach motivates new computational concepts and tools. Here, a fine-grain MIMD computer is emulated and used for the variable region vector quantizer design. Experimental results on digital speech and images are given.

Original languageEnglish
Pages (from-to)36
Number of pages1
JournalNeural Networks
Volume1
Issue number1 SUPPL
DOIs
Publication statusPublished - 1988
Externally publishedYes

Fingerprint

Vector quantization
Information Storage and Retrieval
Processing
Weights and Measures

ASJC Scopus subject areas

  • Artificial Intelligence
  • Neuroscience(all)

Cite this

Vector quantization with optimized grouping and parallel distributed processing. / Matsuyama, Y.

In: Neural Networks, Vol. 1, No. 1 SUPPL, 1988, p. 36.

Research output: Contribution to journalArticle

@article{8bd2c29b6fd1458dac236c42c1b4573b,
title = "Vector quantization with optimized grouping and parallel distributed processing",
abstract = "Vector quantization with optimized grouping of elements is studied. The presented vector quantization allows optimal or suboptimal grouping of source data. Thus, the algorithms herein are called variable region vector quantization. The optimization yielding the data subgroups can also be interpreted as the connection weight adjustmen. The presented methods are still executable on conventional SISD computers. However, the adaptation of the variable region vector quantization to SIMD and MIMD computation via PDP (Parallel Distributed Processing) approach motivates new computational concepts and tools. Here, a fine-grain MIMD computer is emulated and used for the variable region vector quantizer design. Experimental results on digital speech and images are given.",
author = "Y. Matsuyama",
year = "1988",
doi = "10.1016/0893-6080(88)90078-0",
language = "English",
volume = "1",
pages = "36",
journal = "Neural Networks",
issn = "0893-6080",
publisher = "Elsevier Limited",
number = "1 SUPPL",

}

TY - JOUR

T1 - Vector quantization with optimized grouping and parallel distributed processing

AU - Matsuyama, Y.

PY - 1988

Y1 - 1988

N2 - Vector quantization with optimized grouping of elements is studied. The presented vector quantization allows optimal or suboptimal grouping of source data. Thus, the algorithms herein are called variable region vector quantization. The optimization yielding the data subgroups can also be interpreted as the connection weight adjustmen. The presented methods are still executable on conventional SISD computers. However, the adaptation of the variable region vector quantization to SIMD and MIMD computation via PDP (Parallel Distributed Processing) approach motivates new computational concepts and tools. Here, a fine-grain MIMD computer is emulated and used for the variable region vector quantizer design. Experimental results on digital speech and images are given.

AB - Vector quantization with optimized grouping of elements is studied. The presented vector quantization allows optimal or suboptimal grouping of source data. Thus, the algorithms herein are called variable region vector quantization. The optimization yielding the data subgroups can also be interpreted as the connection weight adjustmen. The presented methods are still executable on conventional SISD computers. However, the adaptation of the variable region vector quantization to SIMD and MIMD computation via PDP (Parallel Distributed Processing) approach motivates new computational concepts and tools. Here, a fine-grain MIMD computer is emulated and used for the variable region vector quantizer design. Experimental results on digital speech and images are given.

UR - http://www.scopus.com/inward/record.url?scp=0024175192&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024175192&partnerID=8YFLogxK

U2 - 10.1016/0893-6080(88)90078-0

DO - 10.1016/0893-6080(88)90078-0

M3 - Article

VL - 1

SP - 36

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

IS - 1 SUPPL

ER -