Characteristics of Hebbian-type associative memories with quantized interconnections

Author

Chung, Pau-Choo ; Tsai, Ching-Tsorng ; Sun, Yung-Nien

Author_Institution

Dept. of Electr. Eng., Nat. Cheng Kung Univ., Tainan, Taiwan

Volume

41

Issue

2

fYear

1994

fDate

2/1/1994 12:00:00 AM

Firstpage

168

Lastpage

171

Abstract

The effects of quantized Hebbian-type associative memories (HAM´s) on storage capacity and hardware implementations are explored in this paper. The quantizations include a two-level strategy and a three-level strategy using a cutoff threshold in which the interconnection weights of both strategies are restricted to binary memory points on each synapse. The restricted interconnection weights enable a digital or an analog HAM to be implemented more space efficiently. The three-level quantized network with sparse connectivity among neurons has further simplified analog circuits in the dedicated hardwares. Results of simulation and theory show that the three-level quantized network with a properly selected cutoff threshold possesses higher storage capacity than the two-level quantized network

Keywords

Hebbian learning; Monte Carlo methods; analogue storage; analogue-digital conversion; content-addressable storage; Hebbian-type associative memories; Monte Carlo simulation; analog HAM; binary memory points; cutoff threshold; digital HAM; hardware implementation; interconnection weights; quantized interconnections; simulation; sparse connectivity; storage capacity; synaptic weights; three-level strategy; two-level strategy; Analog circuits; Associative memory; Circuit simulation; Equations; Hardware; Integrated circuit interconnections; Neurodynamics; Neurons; Quantization; Sun;

fLanguage

English

Journal_Title

Circuits and Systems I: Fundamental Theory and Applications, IEEE Transactions on

Publisher

ieee

ISSN

1057-7122

Type

jour

DOI

10.1109/81.269054

Filename

269054