A multi-winner associative memory

Jiongtao Huang, Masafumi Hagiwara

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We propose a new associative memory named Multi-Winner Associative Memory (MWAM) and study its bidirectional association properties in this paper. The proposed MWAM has two processes for pattern pairs storage: storage process and recall process. For the storage process, the proposed MWAM can represent a half of pattern pair in the distributed representation layer and can store the correspondence of pattern and its representation using the upward weights. In addition, the MWAM can store the correspondence of the distributed representation and the other half of pattern pair in the downward weights. For the recall process, the MWAM can recall information bidirectionally: a half of the stored pattern pair can be recalled by receiving the other half in the input-output layer for any stored pattern pairs.

Original languageEnglish
Pages (from-to)1117-1125
Number of pages9
JournalIEICE Transactions on Information and Systems
VolumeE82-D
Issue number7
Publication statusPublished - 1999

Fingerprint

Data storage equipment

Keywords

  • Associative memory
  • B A At
  • Distributed representation
  • Error correction learning
  • Multi-winner

ASJC Scopus subject areas

  • Information Systems
  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

A multi-winner associative memory. / Huang, Jiongtao; Hagiwara, Masafumi.

In: IEICE Transactions on Information and Systems, Vol. E82-D, No. 7, 1999, p. 1117-1125.

Research output: Contribution to journalArticle

Huang, Jiongtao ; Hagiwara, Masafumi. / A multi-winner associative memory. In: IEICE Transactions on Information and Systems. 1999 ; Vol. E82-D, No. 7. pp. 1117-1125.
@article{5e7dfb7b97e34143bb6b4896d57b2e79,
title = "A multi-winner associative memory",
abstract = "We propose a new associative memory named Multi-Winner Associative Memory (MWAM) and study its bidirectional association properties in this paper. The proposed MWAM has two processes for pattern pairs storage: storage process and recall process. For the storage process, the proposed MWAM can represent a half of pattern pair in the distributed representation layer and can store the correspondence of pattern and its representation using the upward weights. In addition, the MWAM can store the correspondence of the distributed representation and the other half of pattern pair in the downward weights. For the recall process, the MWAM can recall information bidirectionally: a half of the stored pattern pair can be recalled by receiving the other half in the input-output layer for any stored pattern pairs.",
keywords = "Associative memory, B A At, Distributed representation, Error correction learning, Multi-winner",
author = "Jiongtao Huang and Masafumi Hagiwara",
year = "1999",
language = "English",
volume = "E82-D",
pages = "1117--1125",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "7",

}

TY - JOUR

T1 - A multi-winner associative memory

AU - Huang, Jiongtao

AU - Hagiwara, Masafumi

PY - 1999

Y1 - 1999

N2 - We propose a new associative memory named Multi-Winner Associative Memory (MWAM) and study its bidirectional association properties in this paper. The proposed MWAM has two processes for pattern pairs storage: storage process and recall process. For the storage process, the proposed MWAM can represent a half of pattern pair in the distributed representation layer and can store the correspondence of pattern and its representation using the upward weights. In addition, the MWAM can store the correspondence of the distributed representation and the other half of pattern pair in the downward weights. For the recall process, the MWAM can recall information bidirectionally: a half of the stored pattern pair can be recalled by receiving the other half in the input-output layer for any stored pattern pairs.

AB - We propose a new associative memory named Multi-Winner Associative Memory (MWAM) and study its bidirectional association properties in this paper. The proposed MWAM has two processes for pattern pairs storage: storage process and recall process. For the storage process, the proposed MWAM can represent a half of pattern pair in the distributed representation layer and can store the correspondence of pattern and its representation using the upward weights. In addition, the MWAM can store the correspondence of the distributed representation and the other half of pattern pair in the downward weights. For the recall process, the MWAM can recall information bidirectionally: a half of the stored pattern pair can be recalled by receiving the other half in the input-output layer for any stored pattern pairs.

KW - Associative memory

KW - B A At

KW - Distributed representation

KW - Error correction learning

KW - Multi-winner

UR - http://www.scopus.com/inward/record.url?scp=0033161875&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033161875&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0033161875

VL - E82-D

SP - 1117

EP - 1125

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 7

ER -