Reliability characteristics of Hebbian-type associative memories in network implementations

Pau Choo Chung, Thomas F. Krile

研究成果: Conference contribution

1 引文 斯高帕斯(Scopus)

摘要

The performance of Hebbian-type associative memories (HAMs) in the presence of faulty interconnections is examined, and equations for predicting network reliability are developed. Optical and VLSI implementations of HAMs are introduced, and the distributions of faulty interconnections in both implementations are discussed. The interconnection faults considered are the equivalent of open-circuit and short-circuit synaptic interconnections. Equations relating the probability of direct one-step convergence (Pdc) to the percentage of failed interconnections are developed for both types of interconnection faults. Monte Carlo simulations indicate that the equations considered here can estimate Pdc accurately. Based on the equations, network performance with failed interconnections can be predicted and trade-offs in network design can be determined before proceeding to implementation. The performance of networks with clustered failed interconnections is also discussed and compared with that of networks with randomly distributed faults. The present results are discussed from the implementation point of view.

原文English
主出版物標題Proceedings. IJCNN - International Joint Conference on Neural Networks
編輯 Anon
發行者Publ by IEEE
頁面363-368
頁數6
ISBN(列印)0780301641
出版狀態Published - 1992 一月 1
事件International Joint Conference on Neural Networks - IJCNN-91-Seattle - Seattle, WA, USA
持續時間: 1991 七月 81991 七月 12

出版系列

名字Proceedings. IJCNN - International Joint Conference on Neural Networks

Other

OtherInternational Joint Conference on Neural Networks - IJCNN-91-Seattle
城市Seattle, WA, USA
期間91-07-0891-07-12

    指紋

All Science Journal Classification (ASJC) codes

  • Engineering(all)

引用此

Chung, P. C., & Krile, T. F. (1992). Reliability characteristics of Hebbian-type associative memories in network implementations. 於 Anon (編輯), Proceedings. IJCNN - International Joint Conference on Neural Networks (頁 363-368). (Proceedings. IJCNN - International Joint Conference on Neural Networks). Publ by IEEE.