Shift-Invariant Associative Memory Based on Homogeneous Neural Networks(<Special Section>Nonlinear Theory and its Applications)
スポンサーリンク
概要
- 論文の詳細を見る
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, (mk)/(2m log m), of patterns, where m is the number of neurons and k is the order of connections.
- 社団法人電子情報通信学会の論文
- 2005-10-01
著者
-
SHIGEI Noritaka
Department of Electrical and Electronics Engineering, Faculty of Engineering, Kagoshima University
-
MIYAJIMA Hiromi
Department of Electrical and Electronics Engineering, Faculty of Engineering, Kagoshima University
-
SHIGEI Noritaka
Kagoshima University
-
MIYAJIMA Hiromi
Kagoshima University
-
Shigei N
Department Of Electrical And Electronics Engineering Faculty Of Engineering Kagoshima University
-
YATSUKI Shuji
Kyoto Software Research, Inc.
-
Yatsuki Shuji
Kyoto Software Research Inc.
-
Miyajima H
Department Of Electrical And Electronics Engineering Faculty Of Engineering Kagoshima University
関連論文
- A Creating Method of Fuzzy Inference Rules by Self-Creating Neural Network
- An Investigation of Fuzzy Model Using AIC
- Constructive, Destructive and Simplified Learning Methods of Fuzzy Inference
- Learning Model in Relaxation Algorithm Influenced by Self-Organizing Maps for Image Restoration
- Numerical Evaluation of Incremental Vector Quantization Using Stochastic Relaxation(Nonlinear Theory and its Applications)
- A Hybrid Learning Approach to Self-Organizing Neural Network for Vector Quantization(Neuro, Fuzzy, GA)(Nonlinear Theory and its Applications)
- Embedding Chordal Rings and Pyramids into Mesh-Connected Computers with Multiple Buses
- Shift-Invariant Associative Memory Based on Homogeneous Neural Networks(Nonlinear Theory and its Applications)
- Decision Procedure of Initial Values of Fuzzy Inference System Using Counterpropagation Networks
- On the Search for Effective Spare Arrangement of Reconfigurable Processor Arrays Using Genetic Algorithm (Special Section on Nonlinear Theory and Its Applications)
- Maximum Finding on One-Way Mesh-Connected Computers with Multiple Buses (Special Section of Papers Selected from ITC-CSCC'96)
- On Efficient Spare Arrangements and an Algorithm with Relocating Spares for Reconfiguring Processor Arrays (Special Section of Papers Selected from ITC-CSCC'96)
- On Methods for Reconfiguring Processor Arrays (Special Issue on Architectures, Algorithms and Networks for Massively parallel Computing)
- On Some Dynamical Properties of Threshold and Homogeneous Networks (Special Section on Nonlinear Theory and Its Applications)
- Reconfiguration Classes and an Optimal Reconfiguration Method within a Reconfiguration Class(Regular Section)