An auto-correlation associative memory which stores plural pattern vectors as its minimum states

Author

Murashima, S. ; Fuchida, Takeshi ; Ida, Tetsuo ; Miyajima, Hiroki

Author_Institution

Dept. of Inf. & Comput. Sci., Kagoshima Univ., Japan

Volume

3

fYear

1993

fDate

25-29 Oct. 1993

Firstpage

2339

Abstract

A noise tolerant auto-correlation associative memory is proposed. An associated energy function is formed by a multiplication of plural Hopfield energy functions each of which includes a 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 around itself. The numerical simulations show a fairly good performance of associative memory for arbitrary pattern vectors which are not orthogonal to each other.

Keywords

content-addressable storage; neural nets; optimisation; asynchronous optimizing algorithm; binary neuron model; energy function; noise tolerant auto-correlation associative memory; pattern vectors; plural Hopfield energy functions; Associative memory; Autocorrelation; Computer science; Equations; Hamming distance; Information retrieval; Neurons; Numerical simulation; Power engineering and energy; Zinc;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on

Print_ISBN

0-7803-1421-2

Type

conf

DOI

10.1109/IJCNN.1993.714194

Filename

714194