Recurrent neural nets as dynamical Boolean systems with application to associative memory

Author

Watta, Paul B. ; Wang, Kaining ; Hassoun, Mohamad H.

Author_Institution

Dept. of Electr. & Comput. Eng., Wayne State Univ., Detroit, MI, USA

Volume

8

Issue

6

fYear

1997

fDate

11/1/1997 12:00:00 AM

Firstpage

1268

Lastpage

1280

Abstract

Discrete-time/discrete-state recurrent neural networks are analyzed from a dynamical Boolean systems point of view in order to devise new analytic and design methods for the class of both single and multilayer recurrent artificial neural networks. With the proposed dynamical Boolean systems analysis, we are able to formulate necessary and sufficient conditions for network stability which are more general than the well-known but restrictive conditions for the class of single layer networks: (1) symmetric weight matrix with (2) positive diagonal and (3) asynchronous update. In terms of design, we use a dynamical Boolean systems analysis to construct a high performance associative memory. With this Boolean memory, we can guarantee that all fundamental memories are stored, and also guarantee the size of the basin of attraction for each fundamental memory

Keywords

Boolean algebra; content-addressable storage; multilayer perceptrons; recurrent neural nets; associative memory; asynchronous update; attraction basin; discrete-state recurrent neural networks; discrete-time recurrent neural networks; dynamical Boolean systems analysis; multilayer neural networks; necessary and sufficient conditions; positive diagonal; symmetric weight matrix; Associative memory; Computer networks; Equations; Laboratories; Lyapunov method; Multi-layer neural network; Neural networks; Recurrent neural networks; Stability analysis; Symmetric matrices;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/72.641450

Filename

641450