On the memorization accuracy of autoassociative memory models

Kazuaki Masuda, Eitaro Aiyoshi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

An autoassociative memory which is modeled as the standard recurrent neural network (N.N.) is capable of storing multiple patterns and subsequently recalling one of them in response to an input signal. However, we found in our recent trials that it can't always recall correct patterns accurately. In this paper, we demonstrate such phenomena by numerical examples and identify the cause of memorization errors. We also propose an immediate solution to memorize correct patterns without fail by storing extra patterns at the same time.

Original languageEnglish
Title of host publicationProceedings of the SICE Annual Conference
Pages530-536
Number of pages7
Publication statusPublished - 2011
Event50th Annual Conference on Society of Instrument and Control Engineers, SICE 2011 - Tokyo, Japan
Duration: 2011 Sep 132011 Sep 18

Other

Other50th Annual Conference on Society of Instrument and Control Engineers, SICE 2011
CountryJapan
CityTokyo
Period11/9/1311/9/18

Fingerprint

Recurrent neural networks
Data storage equipment

Keywords

  • autoassociative memory
  • local optimality
  • nonlinear dynamical system
  • recurrent neural network

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Computer Science Applications

Cite this

Masuda, K., & Aiyoshi, E. (2011). On the memorization accuracy of autoassociative memory models. In Proceedings of the SICE Annual Conference (pp. 530-536). [6060716]

On the memorization accuracy of autoassociative memory models. / Masuda, Kazuaki; Aiyoshi, Eitaro.

Proceedings of the SICE Annual Conference. 2011. p. 530-536 6060716.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Masuda, K & Aiyoshi, E 2011, On the memorization accuracy of autoassociative memory models. in Proceedings of the SICE Annual Conference., 6060716, pp. 530-536, 50th Annual Conference on Society of Instrument and Control Engineers, SICE 2011, Tokyo, Japan, 11/9/13.
Masuda K, Aiyoshi E. On the memorization accuracy of autoassociative memory models. In Proceedings of the SICE Annual Conference. 2011. p. 530-536. 6060716
Masuda, Kazuaki ; Aiyoshi, Eitaro. / On the memorization accuracy of autoassociative memory models. Proceedings of the SICE Annual Conference. 2011. pp. 530-536
@inproceedings{15d82645edfb4b9489e2b9d900215c0b,
title = "On the memorization accuracy of autoassociative memory models",
abstract = "An autoassociative memory which is modeled as the standard recurrent neural network (N.N.) is capable of storing multiple patterns and subsequently recalling one of them in response to an input signal. However, we found in our recent trials that it can't always recall correct patterns accurately. In this paper, we demonstrate such phenomena by numerical examples and identify the cause of memorization errors. We also propose an immediate solution to memorize correct patterns without fail by storing extra patterns at the same time.",
keywords = "autoassociative memory, local optimality, nonlinear dynamical system, recurrent neural network",
author = "Kazuaki Masuda and Eitaro Aiyoshi",
year = "2011",
language = "English",
isbn = "9784907764395",
pages = "530--536",
booktitle = "Proceedings of the SICE Annual Conference",

}

TY - GEN

T1 - On the memorization accuracy of autoassociative memory models

AU - Masuda, Kazuaki

AU - Aiyoshi, Eitaro

PY - 2011

Y1 - 2011

N2 - An autoassociative memory which is modeled as the standard recurrent neural network (N.N.) is capable of storing multiple patterns and subsequently recalling one of them in response to an input signal. However, we found in our recent trials that it can't always recall correct patterns accurately. In this paper, we demonstrate such phenomena by numerical examples and identify the cause of memorization errors. We also propose an immediate solution to memorize correct patterns without fail by storing extra patterns at the same time.

AB - An autoassociative memory which is modeled as the standard recurrent neural network (N.N.) is capable of storing multiple patterns and subsequently recalling one of them in response to an input signal. However, we found in our recent trials that it can't always recall correct patterns accurately. In this paper, we demonstrate such phenomena by numerical examples and identify the cause of memorization errors. We also propose an immediate solution to memorize correct patterns without fail by storing extra patterns at the same time.

KW - autoassociative memory

KW - local optimality

KW - nonlinear dynamical system

KW - recurrent neural network

UR - http://www.scopus.com/inward/record.url?scp=81255147443&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=81255147443&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:81255147443

SN - 9784907764395

SP - 530

EP - 536

BT - Proceedings of the SICE Annual Conference

ER -