Applications of a connection code computer for neural network calculations

Author

Morishta, Takayuki ; Hatano, Fumihiro ; Teramoto, Iwao

Author_Institution

Fac. of Comput. Sci. & Syst. Eng., Okayama Prefectural Univ., Japan

Volume

1

fYear

1998

fDate

4-8 May 1998

Firstpage

609

Abstract

A model of a connection code computer, which we proposed as a reconfigurable hardware, is developed for modelling neural networks. This hardware is able to speed up the parallel and pipelined calculations by realizing the optimal composition (hardware structure) for various problems. As an example, one of typical neural network models, the backpropagation model, is simulated. Each component is improved to reduce the chip-size. All components are designed and are implemented to check the sizes and their performance

Keywords

backpropagation; neural nets; parallel architectures; pipeline processing; reconfigurable architectures; simulation; backpropagation model; connection code computer; neural network models; parallel architecture; parallel processing; pipeline processing; reconfigurable hardware; Application software; Computer architecture; Computer networks; Computer science; Concurrent computing; Gold; Neural network hardware; Neural networks; Pipeline processing; Systems engineering and theory;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on

Conference_Location

Anchorage, AK

ISSN

1098-7576

Print_ISBN

0-7803-4859-1

Type

conf

DOI

10.1109/IJCNN.1998.682348

Filename

682348