A Neural Architecture for Content as well as Address-based Storage and Recall: Theory and Applications

Chun HSIEN Chen, Vasant Honavar

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper presents an approach to design of a neural architecture for both associative (content-addressed) and address-based memories. Several interesting properties of the memory module are mathematically analyzed in detail. When used as an associative memory, the proposed neural memory module supports recall from partial input patterns, (sequential) multiple recalls and fault-tolerance. When used as an address-based memory, the memory module can provide working space for dynamic representations for symbol processing and shared message-passing among neural network modules within an integrated neural network system. It also provides for real-time update of memory contents by one-shot learning without interference with other stored patterns.

Original languageEnglish (US)
Pages (from-to)281-300
Number of pages20
JournalConnection Science
Volume7
Issue number3-4
DOIs
StatePublished - Sep 1 1995

Fingerprint

Data storage equipment
Neural networks
Message passing
Fault tolerance
Processing

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Artificial Intelligence

Cite this

@article{8e3c74290310464eb74b0b46f764f945,
title = "A Neural Architecture for Content as well as Address-based Storage and Recall: Theory and Applications",
abstract = "This paper presents an approach to design of a neural architecture for both associative (content-addressed) and address-based memories. Several interesting properties of the memory module are mathematically analyzed in detail. When used as an associative memory, the proposed neural memory module supports recall from partial input patterns, (sequential) multiple recalls and fault-tolerance. When used as an address-based memory, the memory module can provide working space for dynamic representations for symbol processing and shared message-passing among neural network modules within an integrated neural network system. It also provides for real-time update of memory contents by one-shot learning without interference with other stored patterns.",
author = "Chen, {Chun HSIEN} and Vasant Honavar",
year = "1995",
month = "9",
day = "1",
doi = "10.1080/09540099550039255",
language = "English (US)",
volume = "7",
pages = "281--300",
journal = "Connection Science",
issn = "0954-0091",
publisher = "Taylor and Francis AS",
number = "3-4",

}

A Neural Architecture for Content as well as Address-based Storage and Recall : Theory and Applications. / Chen, Chun HSIEN; Honavar, Vasant.

In: Connection Science, Vol. 7, No. 3-4, 01.09.1995, p. 281-300.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Neural Architecture for Content as well as Address-based Storage and Recall

T2 - Theory and Applications

AU - Chen, Chun HSIEN

AU - Honavar, Vasant

PY - 1995/9/1

Y1 - 1995/9/1

N2 - This paper presents an approach to design of a neural architecture for both associative (content-addressed) and address-based memories. Several interesting properties of the memory module are mathematically analyzed in detail. When used as an associative memory, the proposed neural memory module supports recall from partial input patterns, (sequential) multiple recalls and fault-tolerance. When used as an address-based memory, the memory module can provide working space for dynamic representations for symbol processing and shared message-passing among neural network modules within an integrated neural network system. It also provides for real-time update of memory contents by one-shot learning without interference with other stored patterns.

AB - This paper presents an approach to design of a neural architecture for both associative (content-addressed) and address-based memories. Several interesting properties of the memory module are mathematically analyzed in detail. When used as an associative memory, the proposed neural memory module supports recall from partial input patterns, (sequential) multiple recalls and fault-tolerance. When used as an address-based memory, the memory module can provide working space for dynamic representations for symbol processing and shared message-passing among neural network modules within an integrated neural network system. It also provides for real-time update of memory contents by one-shot learning without interference with other stored patterns.

UR - http://www.scopus.com/inward/record.url?scp=33747670688&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33747670688&partnerID=8YFLogxK

U2 - 10.1080/09540099550039255

DO - 10.1080/09540099550039255

M3 - Article

AN - SCOPUS:33747670688

VL - 7

SP - 281

EP - 300

JO - Connection Science

JF - Connection Science

SN - 0954-0091

IS - 3-4

ER -