An Auto-Correlation Associative Memory which Has an Energy Function of Higher Order
Summary :
A noise tolerant auto-correlation associative memory is proposed. An associated energy function is formed by a multiplication of plural Hopfield's energy functions each of which includes single pattern as its energy minimum. An asynchronous optimizing algorithm of the whole energy function is also presented based on the binary neuron model. The advantages of this new associative memory are that the orthogonality relation among patterns does not need to be satisfied and each stored pattern has a large basin of attraction around itself. The computer simulations show a fairly good performance of associative memory for arbitrary pattern vectors which are not orthogonal to each other.
- Publication
- IEICE TRANSACTIONS on Fundamentals Vol.E78-A No.3 pp.424-430
- Publication Date
- 1995/03/25
- Publicized
- Online ISSN
- DOI
- Type of Manuscript
- PAPER
- Category
- Neural Networks
Authors
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Sadayuki MURASHIMA, Takayasu FUCHIDA, Toshihiro IDA, Takayuki TOYOHIRA, Hiromi MIYAJIMA, "An Auto-Correlation Associative Memory which Has an Energy Function of Higher Order" in IEICE TRANSACTIONS on Fundamentals,
vol. E78-A, no. 3, pp. 424-430, March 1995, doi: .
Abstract: A noise tolerant auto-correlation associative memory is proposed. An associated energy function is formed by a multiplication of plural Hopfield's energy functions each of which includes single pattern as its energy minimum. An asynchronous optimizing algorithm of the whole energy function is also presented based on the binary neuron model. The advantages of this new associative memory are that the orthogonality relation among patterns does not need to be satisfied and each stored pattern has a large basin of attraction around itself. The computer simulations show a fairly good performance of associative memory for arbitrary pattern vectors which are not orthogonal to each other.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e78-a_3_424/_p
Copy
@ARTICLE{e78-a_3_424,
author={Sadayuki MURASHIMA, Takayasu FUCHIDA, Toshihiro IDA, Takayuki TOYOHIRA, Hiromi MIYAJIMA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={An Auto-Correlation Associative Memory which Has an Energy Function of Higher Order},
year={1995},
volume={E78-A},
number={3},
pages={424-430},
abstract={A noise tolerant auto-correlation associative memory is proposed. An associated energy function is formed by a multiplication of plural Hopfield's energy functions each of which includes single pattern as its energy minimum. An asynchronous optimizing algorithm of the whole energy function is also presented based on the binary neuron model. The advantages of this new associative memory are that the orthogonality relation among patterns does not need to be satisfied and each stored pattern has a large basin of attraction around itself. The computer simulations show a fairly good performance of associative memory for arbitrary pattern vectors which are not orthogonal to each other.},
keywords={},
doi={},
ISSN={},
month={March},}
Copy
TY - JOUR
TI - An Auto-Correlation Associative Memory which Has an Energy Function of Higher Order
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 424
EP - 430
AU - Sadayuki MURASHIMA
AU - Takayasu FUCHIDA
AU - Toshihiro IDA
AU - Takayuki TOYOHIRA
AU - Hiromi MIYAJIMA
PY - 1995
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E78-A
IS - 3
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - March 1995
AB - A noise tolerant auto-correlation associative memory is proposed. An associated energy function is formed by a multiplication of plural Hopfield's energy functions each of which includes single pattern as its energy minimum. An asynchronous optimizing algorithm of the whole energy function is also presented based on the binary neuron model. The advantages of this new associative memory are that the orthogonality relation among patterns does not need to be satisfied and each stored pattern has a large basin of attraction around itself. The computer simulations show a fairly good performance of associative memory for arbitrary pattern vectors which are not orthogonal to each other.
ER -
English
