Newsgroups: comp.ai.neural-nets,comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!travelers.mail.cornell.edu!news.kei.com!news.mathworks.com!tank.news.pipex.net!pipex!usenet.eel.ufl.edu!usenet.cis.ufl.edu!usenet.ufl.edu!zeno.fit.edu!zach.fit.edu!ram82001
From: ram82001@zach.fit.edu (Gary Russel /ADVISOR L. FAUSETT)
Subject: Re: Tasks of Thought
Message-ID: <DF7Jt3.6ME@zeno.fit.edu>
Sender: news@zeno.fit.edu (USENET NEWS SYSTEM)
Nntp-Posting-Host: zach.fit.edu
Organization: Florida Institute of Technology
References: <DEqoFM.35K@festival.ed.ac.uk> <moran-110995172424@gb22.carswell.wfu.edu> <43n3u8$37v@adapter.src.umd.edu>
Date: Wed, 20 Sep 1995 14:19:51 GMT
Lines: 19
Xref: glinda.oz.cs.cmu.edu comp.ai.neural-nets:26991 comp.ai.philosophy:32872

In article <43n3u8$37v@adapter.src.umd.edu> delliott@Glue.umd.edu (David L. Elliott) writes:
>In article <moran-110995172424@gb22.carswell.wfu.edu>,
>Patrick Moran <moran@wfu.edu> wrote:
>>In article <DEqoFM.35K@festival.ed.ac.uk>, cam@castle.ed.ac.uk (Chris
>>Malcolm) wrote:
>>
>>> In my school chess club we had a natural chess genius who played very
>>> fast, and who was *very* hard to beat. The odd thing was that he could
>>> never explain why he made his moves, they just "felt right" in some
>>> intuitive way. ....

What was the competition like?  Were his opponents required to play
quickly also (speed chess)?

In the work I've done with RBF lately, the (random or arbitrary) values
for centers, weights, and spreads create a very large error.  After just
*one* epoch, the error is greatly reduced but still significant.
Likewise, this chess player may be finding good moves but not the best
moves (or maybe "likewise" isn't the right word).
