Why VLSI implementations of associative VLCNs require connection multiplexing
- 1 January 1988
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 173-180 vol.2
- https://doi.org/10.1109/icnn.1988.23926
Abstract
A discussion is presented of some of the implementation constraints imposed on VLSI architectures for emulations of very large connectionist/neural networks (VLCNs). Specifically, the authors show that multiplexing of interconnections is necessary for networks exhibiting poor locality. They show that it is more feasible to build a VLCN system with sharing or multiplexing of interconnections than to build one with dedicated wires for each connection. This is true unless the network exhibits extreme locality and all CNs are connected to others within some small-radius region. Unfortunately, association requires some global connectivity.<>Keywords
This publication has 3 references indexed in Scilit:
- A silicon model of early visual processingNeural Networks, 1988
- A CMOS associative memory chip based on neural networksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1987
- Artificial neural networks for computingJournal of Vacuum Science & Technology B, 1986