Multi-valued associative memory neural network

Chun Bo Xiu; Xiang Dong Liu; Yu He Zhang

Multi-valued associative memory neural network

Chun Bo Xiu^*, Xiang Dong Liu, Yu He Zhang

^*Corresponding author for this work

School of Automation

Beijing Institute of Technology

Research output: Contribution to journal › Article › peer-review

Abstract

A learning method for multi-valued associative memory network is introduced, which is based on Hebb rule but utilizes more information. According to the current probe vector, the connection weights matrix can be chosen dynamically. Double-valued associative memory and multi-valued associative memory are all realized in our simulation experiment. The experimental results show that the method can enhance the associative success rate.

Original language	English
Pages (from-to)	352-356
Number of pages	5
Journal	Journal of Beijing Institute of Technology (English Edition)
Volume	12
Issue number	4
Publication status	Published - Dec 2003

Keywords

Associative memory
Gray-scale images
Learning method
Neural network

Cite this

@article{6d77cad618aa4bafa1b5b502ae62c228,

title = "Multi-valued associative memory neural network",

abstract = "A learning method for multi-valued associative memory network is introduced, which is based on Hebb rule but utilizes more information. According to the current probe vector, the connection weights matrix can be chosen dynamically. Double-valued associative memory and multi-valued associative memory are all realized in our simulation experiment. The experimental results show that the method can enhance the associative success rate.",

keywords = "Associative memory, Gray-scale images, Learning method, Neural network",

author = "Xiu, {Chun Bo} and Liu, {Xiang Dong} and Zhang, {Yu He}",

year = "2003",

month = dec,

language = "English",

volume = "12",

pages = "352--356",

journal = "Journal of Beijing Institute of Technology (English Edition)",

issn = "1004-0579",

publisher = "Beijing Institute of Technology",

number = "4",

}

TY - JOUR

T1 - Multi-valued associative memory neural network

AU - Xiu, Chun Bo

AU - Liu, Xiang Dong

AU - Zhang, Yu He

PY - 2003/12

Y1 - 2003/12

N2 - A learning method for multi-valued associative memory network is introduced, which is based on Hebb rule but utilizes more information. According to the current probe vector, the connection weights matrix can be chosen dynamically. Double-valued associative memory and multi-valued associative memory are all realized in our simulation experiment. The experimental results show that the method can enhance the associative success rate.

AB - A learning method for multi-valued associative memory network is introduced, which is based on Hebb rule but utilizes more information. According to the current probe vector, the connection weights matrix can be chosen dynamically. Double-valued associative memory and multi-valued associative memory are all realized in our simulation experiment. The experimental results show that the method can enhance the associative success rate.

KW - Associative memory

KW - Gray-scale images

KW - Learning method

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=2942525280&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:2942525280

SN - 1004-0579

VL - 12

SP - 352

EP - 356

JO - Journal of Beijing Institute of Technology (English Edition)

JF - Journal of Beijing Institute of Technology (English Edition)

IS - 4

ER -

Multi-valued associative memory neural network

Abstract

Keywords

Other files and links

Fingerprint

Cite this