Implementation of an oversize neural network on DAP-510

Author

Gupta, S.N. ; Zubair, M. ; Grosch, C.E.

Author_Institution

Dept. of Comput. Sci., Old Dominion Univ., Norfolk, VA, USA

fYear

1991

fDate

30 Apr-2 May 1991

Firstpage

161

Lastpage

164

Abstract

The authors propose parallel implementation of a multi-layer feedforward neural network on the AMT DAP-510. The network is trained using the Back Propagation algorithm. The proposed implementation is for an oversize network, that is for a network which has more numbers of neurons than the processors. Also, the implementation has the flexibility of varying the size of the network. The peak performance of 2.6 million interconnections per second is achieved for a three-layer network with 512 neurons in each layer. They compare their results with the existing parallel implementations

Keywords

neural nets; parallel processing; performance evaluation; virtual machines; 2.6 million interconnections per second; 512 neurons; AMT DAP-510; Back Propagation algorithm; multi-layer feedforward neural network; oversize neural network; three-layer network; Computational modeling; Computer networks; Computer science; Computer simulation; Concurrent computing; Feedforward neural networks; Feedforward systems; Multi-layer neural network; Neural networks; Neurons;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel Processing Symposium, 1991. Proceedings., Fifth International

Conference_Location

Anaheim, CA

Print_ISBN

0-8186-9167-0

Type

conf

DOI

10.1109/IPPS.1991.153773

Filename

153773