home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!usc!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!caen!destroyer!cs.ubc.ca!unixg.ubc.ca!kakwa.ucs.ualberta.ca!access.usask.ca!skorpio!choy
- From: choy@skorpio.usask.ca (I am a terminator.)
- Subject: Algorithmic equivalents and performance?
- Message-ID: <1992Nov20.193409.27335@access.usask.ca>
- Sender: choy@skorpio (I am a terminator.)
- Nntp-Posting-Host: skorpio.usask.ca
- Organization: University of Saskatchewan, Saskatoon, Canada
- Date: Fri, 20 Nov 1992 19:34:09 GMT
- Lines: 8
-
- A neural network is an iterative architecture composed of simple
- I/O devices. How would a neural net compare in performance to a
- non-neural net implementation of the neural net's "algorithm"?
- Some kind of algorithmic (likely parallel) implementation. Is there
- a lot of room of optimization?
-
- Henry Choy
- choy@cs.usask.ca
-