Share Email Print

Proceedings Paper

Back Propagation On Warp
Author(s): George L. Gusciora
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

We describe two fast back-propagation algorithms for a linear array of processors. Results of the implementation of both algorithms on Warp, a ten processor, programmable systolic array computer, are reviewed and compared with back-propagation implementations on other machines. Our fastest Warp simulator is about 11 times faster at simulating the NETtaik text-to-speech network than the fastest back-propagation simulator previously reported in the literature. This fast simulator on Warp is being used routinely in autonomous navigation experiments, image processing experiments and speech recognition experiments at Carnegie Mellon. Our results indicate that linear systolic array machines can be efficient neural network simulators.

Paper Details

Date Published: 16 December 1989
PDF: 9 pages
Proc. SPIE 0977, Real-Time Signal Processing XI, (16 December 1989); doi: 10.1117/12.948568
Show Author Affiliations
George L. Gusciora, Carnegie Mellon University (United States)

Published in SPIE Proceedings Vol. 0977:
Real-Time Signal Processing XI
J. P. Letellier, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?