Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech!swrinde!cs.utexas.edu!atlantis.utmb.edu!news.tamu.edu!news.utdallas.edu!corpgate!bcarh189.bnr.ca!nott!cunews!freenet.carleton.ca!FreeNet.Carleton.CA!al255
From: al255@FreeNet.Carleton.CA (Arvon Erickson)
Subject: Re: Othello
Message-ID: <D7KBwA.9Ju@freenet.carleton.ca>
Sender: al255@freenet3.carleton.ca (Arvon Erickson)
Reply-To: al255@FreeNet.Carleton.CA (Arvon Erickson)
Organization: The National Capital FreeNet
References: <PETER.95Apr23123122@swamp.indigo.co.il>  
Date: Mon, 24 Apr 1995 23:28:58 GMT
Lines: 24

In a previous posting, Peter Gordon (peter@swamp.indigo.co.il) writes:
> I would like to write Othello using a neural network.
> 
> Both backprop and recurrent backprop seem unsuitable. I want 
> the network to learn from complete game examples. I don't want 
> to impose external ideas of what is a 'good' position and what 
> is a 'bad' position. The only objective think you can say is who 
> is the winner when the end of the game is reached.
> 
> Does anyone have any ideas?
> 

See the FAQ of rec.games.backgammon.  G. Tesauro wrote a famous
NN backgammon program and you will find references there, one
very recent (in IEEE Computers????  June issue???).  The kind
of NN he used should work like you want for Othello too.



--
A. Erickson
al255@Freenet.carleton.ca


