home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!caen!destroyer!ubc-cs!unixg.ubc.ca!kakwa.ucs.ualberta.ca!alberta!arms
- From: arms@cs.UAlberta.CA (Bill Armstrong)
- Subject: Re: Reducing Training time vs Generalisation
- Message-ID: <arms.714517138@spedden>
- Sender: news@cs.UAlberta.CA (News Administrator)
- Nntp-Posting-Host: spedden.cs.ualberta.ca
- Organization: University of Alberta, Edmonton, Canada
- References: <Bt9GIx.9In.1@cs.cmu.edu> <arms.714289771@spedden> <?.714342847@tazdevil>
- Date: Sat, 22 Aug 1992 20:58:58 GMT
- Lines: 28
-
- henrik@mpci.llnl.gov (Henrik Klagges) writes:
-
- >arms@cs.UAlberta.CA (Bill Armstrong) writes:
-
- ...
-
- >>coming closer all the time to a logical net. If you are able to
- >>replace sigmoids with sharp thresholds, and not change the output of
- >>the net significantly, then you are really using threshold *logic*
- >>nets.
-
- >Well, if this replaceability is there ... would be great ! Don't
- >think so, though. I'd need a few more experiments on that.
- >An inital look suggests that the weights & sigmoids cannot easily
- >(straightforwardly) replaced with 'gates'.
-
- For training, maybe you have a problem. ALNs don't. So we can train
- to get piecewise constant functions. The final step is smoothing,
- which is fairly straightforward, if not trivial, and is efficient.
-
- Bill
-
-
- --
- ***************************************************
- Prof. William W. Armstrong, Computing Science Dept.
- University of Alberta; Edmonton, Alberta, Canada T6G 2H1
- arms@cs.ualberta.ca Tel(403)492 2374 FAX 492 1071
-