Shift-Invariant Associative Memory Based on Homogeneous Neural Networks

Hiromi MIYAJIMA; Noritaka SHIGEI; Shuji YATSUKI

doi:10.1093/ietfec/e88-a.10.2600

Shift-Invariant Associative Memory Based on Homogeneous Neural Networks

Hiromi MIYAJIMA, Noritaka SHIGEI, Shuji YATSUKI

Full Text Views

0

Share
Cite this

Summary :

This paper proposes homogeneous neural networks (HNNs), in which each neuron has identical weights. HNNs can realize shift-invariant associative memory, that is, HNNs can associate not only a memorized pattern but also its shifted ones. The transition property of HNNs is analyzed by the statistical method. We show the probability that each neuron outputs correctly and the error-correcting ability. Further, we show that HNNs cannot memorize over the number,, of patterns, where m is the number of neurons and k is the order of connections.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E88-A No.10 pp.2600-2606

Publication Date: 2005/10/01

Publicized

Online ISSN

DOI: 10.1093/ietfec/e88-a.10.2600

Type of Manuscript: Special Section PAPER (Special Section on Nonlinear Theory and its Applications)

Category

Cite this

Copy

Hiromi MIYAJIMA, Noritaka SHIGEI, Shuji YATSUKI, "Shift-Invariant Associative Memory Based on Homogeneous Neural Networks" in IEICE TRANSACTIONS on Fundamentals, vol. E88-A, no. 10, pp. 2600-2606, October 2005, doi: 10.1093/ietfec/e88-a.10.2600.
Abstract: This paper proposes homogeneous neural networks (HNNs), in which each neuron has identical weights. HNNs can realize shift-invariant associative memory, that is, HNNs can associate not only a memorized pattern but also its shifted ones. The transition property of HNNs is analyzed by the statistical method. We show the probability that each neuron outputs correctly and the error-correcting ability. Further, we show that HNNs cannot memorize over the number,, of patterns, where m is the number of neurons and k is the order of connections.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1093/ietfec/e88-a.10.2600/_p

Copy

@ARTICLE{e88-a_10_2600,
author={Hiromi MIYAJIMA, Noritaka SHIGEI, Shuji YATSUKI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Shift-Invariant Associative Memory Based on Homogeneous Neural Networks},
year={2005},
volume={E88-A},
number={10},
pages={2600-2606},
abstract={This paper proposes homogeneous neural networks (HNNs), in which each neuron has identical weights. HNNs can realize shift-invariant associative memory, that is, HNNs can associate not only a memorized pattern but also its shifted ones. The transition property of HNNs is analyzed by the statistical method. We show the probability that each neuron outputs correctly and the error-correcting ability. Further, we show that HNNs cannot memorize over the number,, of patterns, where m is the number of neurons and k is the order of connections.},
keywords={},
doi={10.1093/ietfec/e88-a.10.2600},
ISSN={},
month={October},}

Copy

TY - JOUR
TI - Shift-Invariant Associative Memory Based on Homogeneous Neural Networks
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2600
EP - 2606
AU - Hiromi MIYAJIMA
AU - Noritaka SHIGEI
AU - Shuji YATSUKI
PY - 2005
DO - 10.1093/ietfec/e88-a.10.2600
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E88-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 2005
AB - This paper proposes homogeneous neural networks (HNNs), in which each neuron has identical weights. HNNs can realize shift-invariant associative memory, that is, HNNs can associate not only a memorized pattern but also its shifted ones. The transition property of HNNs is analyzed by the statistical method. We show the probability that each neuron outputs correctly and the error-correcting ability. Further, we show that HNNs cannot memorize over the number,, of patterns, where m is the number of neurons and k is the order of connections.
ER -