Parallel Algorithms for Neural Networks

CNAPS Server
Sisal implementations

This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

Back to Chuck Anderson's Home Page

CNAPS Server

We have purchased a 128-processor, CNAPS Server, from Adaptive Solutions, Inc.. Currently we use this machine with Adaptive Solutions' Buildnet library of backpropagation routines to perform our EEG experiments. For details of this use and the resulting speedup (4 minutes to perform an experiment versus 16 hours on a Sun Sparc 1), see:

C. Anderson, S. Devulapalli, and E. Stolz. Determining Mental State from EEG Signals Using Neural Networks. Scientific Programming, Special Issue on Applications Analysis, 4, 3, Fall, 1995, 171--183. (0.1 MB compressed postscript)

Back to Contents

Sisal

We are also experimenting with Sisal implementations of neural networks. The "mixture of experts" paradigm automatically decomposes the solution of problems into parallel structures. Using Sisal implementations, we hope to show how this results in automatic distribution of a problem's solution over multiple processors.

Back to Contents

Parallel Algorithms for Neural Networks Research in CS at CSU, Charles W. Anderson / anderson@cs.colostate.edu