Multimodule associative memory for many-to-many associations

Motonobu Hattori, Masafumi Hagiwara

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

In this paper, we propose a novel associative memory model called MultiModule Associative memory for Many-to-Many Associations, (MMA)2 for short. The proposed (MMA)2 consists of multiple modules and each module has a Hopfield type of associative memory. In the (MMA)2, a memory item is regarded as several divided patterns. Each pattern is assigned to each module and the patterns are related to each other in the learning. Unlike a single- layered conventional associative memory, the (MMA)2 can recall a complete memory item from even a single part of it owing to the multiple modules structure. Even if a part of a memory item that is common to the other memory items is given to the proposed (MMA)2, all items that relate to the input can be recalled: that is, the proposed (MMA)2 can deal with the set of memory items which includes one-to-many relations and many-to-many relations such as (A1,B1,C1, ....), (A1,B2,C2 ....),(A2,B2,C3 ....) ..... In order to memorize and recall such very complicated training data, the (MMA)2 employs pseudo-noise (PN) patterns, transformation of distributed patterns into locally represented patterns and the logical operations. These techniques contribute to avoid producing a mixed unknown pattern, which consists of a superimposed pattern of some stored patterns and the cross- talk noise and interferes with recalling correct patterns. A number of computer simulation results show the effectiveness of the proposed (MMA)2. Furthermore, we show that the (MMA)2 can deal with a knowledge processing.

Original languageEnglish
Pages (from-to)99-119
Number of pages21
JournalNeurocomputing
Volume19
Issue number1-3
DOIs
Publication statusPublished - 1998 Apr 21

Fingerprint

Data storage equipment
Noise
Crosstalk
varespladib methyl
Computer Simulation
Learning
Computer simulation
Processing

Keywords

  • Locally represented pattern
  • Many-to-Many Associations
  • Pseudo-Noise (PN) pattern

ASJC Scopus subject areas

  • Artificial Intelligence
  • Cellular and Molecular Neuroscience

Cite this

Multimodule associative memory for many-to-many associations. / Hattori, Motonobu; Hagiwara, Masafumi.

In: Neurocomputing, Vol. 19, No. 1-3, 21.04.1998, p. 99-119.

Research output: Contribution to journalArticle

Hattori, Motonobu ; Hagiwara, Masafumi. / Multimodule associative memory for many-to-many associations. In: Neurocomputing. 1998 ; Vol. 19, No. 1-3. pp. 99-119.
@article{87e0bd6510c649eda820e6e7b988176d,
title = "Multimodule associative memory for many-to-many associations",
abstract = "In this paper, we propose a novel associative memory model called MultiModule Associative memory for Many-to-Many Associations, (MMA)2 for short. The proposed (MMA)2 consists of multiple modules and each module has a Hopfield type of associative memory. In the (MMA)2, a memory item is regarded as several divided patterns. Each pattern is assigned to each module and the patterns are related to each other in the learning. Unlike a single- layered conventional associative memory, the (MMA)2 can recall a complete memory item from even a single part of it owing to the multiple modules structure. Even if a part of a memory item that is common to the other memory items is given to the proposed (MMA)2, all items that relate to the input can be recalled: that is, the proposed (MMA)2 can deal with the set of memory items which includes one-to-many relations and many-to-many relations such as (A1,B1,C1, ....), (A1,B2,C2 ....),(A2,B2,C3 ....) ..... In order to memorize and recall such very complicated training data, the (MMA)2 employs pseudo-noise (PN) patterns, transformation of distributed patterns into locally represented patterns and the logical operations. These techniques contribute to avoid producing a mixed unknown pattern, which consists of a superimposed pattern of some stored patterns and the cross- talk noise and interferes with recalling correct patterns. A number of computer simulation results show the effectiveness of the proposed (MMA)2. Furthermore, we show that the (MMA)2 can deal with a knowledge processing.",
keywords = "Locally represented pattern, Many-to-Many Associations, Pseudo-Noise (PN) pattern",
author = "Motonobu Hattori and Masafumi Hagiwara",
year = "1998",
month = "4",
day = "21",
doi = "10.1016/S0925-2312(97)00100-8",
language = "English",
volume = "19",
pages = "99--119",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",
number = "1-3",

}

TY - JOUR

T1 - Multimodule associative memory for many-to-many associations

AU - Hattori, Motonobu

AU - Hagiwara, Masafumi

PY - 1998/4/21

Y1 - 1998/4/21

N2 - In this paper, we propose a novel associative memory model called MultiModule Associative memory for Many-to-Many Associations, (MMA)2 for short. The proposed (MMA)2 consists of multiple modules and each module has a Hopfield type of associative memory. In the (MMA)2, a memory item is regarded as several divided patterns. Each pattern is assigned to each module and the patterns are related to each other in the learning. Unlike a single- layered conventional associative memory, the (MMA)2 can recall a complete memory item from even a single part of it owing to the multiple modules structure. Even if a part of a memory item that is common to the other memory items is given to the proposed (MMA)2, all items that relate to the input can be recalled: that is, the proposed (MMA)2 can deal with the set of memory items which includes one-to-many relations and many-to-many relations such as (A1,B1,C1, ....), (A1,B2,C2 ....),(A2,B2,C3 ....) ..... In order to memorize and recall such very complicated training data, the (MMA)2 employs pseudo-noise (PN) patterns, transformation of distributed patterns into locally represented patterns and the logical operations. These techniques contribute to avoid producing a mixed unknown pattern, which consists of a superimposed pattern of some stored patterns and the cross- talk noise and interferes with recalling correct patterns. A number of computer simulation results show the effectiveness of the proposed (MMA)2. Furthermore, we show that the (MMA)2 can deal with a knowledge processing.

AB - In this paper, we propose a novel associative memory model called MultiModule Associative memory for Many-to-Many Associations, (MMA)2 for short. The proposed (MMA)2 consists of multiple modules and each module has a Hopfield type of associative memory. In the (MMA)2, a memory item is regarded as several divided patterns. Each pattern is assigned to each module and the patterns are related to each other in the learning. Unlike a single- layered conventional associative memory, the (MMA)2 can recall a complete memory item from even a single part of it owing to the multiple modules structure. Even if a part of a memory item that is common to the other memory items is given to the proposed (MMA)2, all items that relate to the input can be recalled: that is, the proposed (MMA)2 can deal with the set of memory items which includes one-to-many relations and many-to-many relations such as (A1,B1,C1, ....), (A1,B2,C2 ....),(A2,B2,C3 ....) ..... In order to memorize and recall such very complicated training data, the (MMA)2 employs pseudo-noise (PN) patterns, transformation of distributed patterns into locally represented patterns and the logical operations. These techniques contribute to avoid producing a mixed unknown pattern, which consists of a superimposed pattern of some stored patterns and the cross- talk noise and interferes with recalling correct patterns. A number of computer simulation results show the effectiveness of the proposed (MMA)2. Furthermore, we show that the (MMA)2 can deal with a knowledge processing.

KW - Locally represented pattern

KW - Many-to-Many Associations

KW - Pseudo-Noise (PN) pattern

UR - http://www.scopus.com/inward/record.url?scp=0345267232&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0345267232&partnerID=8YFLogxK

U2 - 10.1016/S0925-2312(97)00100-8

DO - 10.1016/S0925-2312(97)00100-8

M3 - Article

AN - SCOPUS:0345267232

VL - 19

SP - 99

EP - 119

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

IS - 1-3

ER -