Newsgroups: comp.ai.genetic,comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!newshost.marcam.com!zip.eecs.umich.edu!caen!msunews!harbinger.cc.monash.edu.au!news.cs.su.oz.au!metro!unsw.edu.au!sserve!canth!darwen
From: darwen@canth.cs.adfa.oz.au (Paul Darwen)
Subject: Problem wanted
Message-ID: <1995Feb14.045434.2683@sserve.cc.adfa.oz.au>
Sender: darwen@canth (Paul Darwen)
Organization: Australian Defence Force Academy
Date: Tue, 14 Feb 1995 04:54:34 GMT
Lines: 40
Xref: glinda.oz.cs.cmu.edu comp.ai.genetic:4954 comp.ai.neural-nets:21985

I am looking for a problem to solve with a machine learning 
system.  This system is a co-evolutionary genetic algorithm, 
similar to that used by Lindgren [1]. 

The problem to test it would ideally be a multi-player game
in which we already know that there are several distinct 
high-quality strategies.  We also require that the game does 
not force players into particular strategies, such as 
an evader-pursuer problem. 

For example, the game of Othello [2] (a.k.a. Reversi) is known 
to have the positional strategy and the more advanced mobility 
strategy.  

Chess and Go are a little on the too-hard side, and are a last 
resort.  Othello and iterated Prisoner's Dilemma have already 
been well-studied.  What other multi-player games are known to 
have distinct high-quality strategies that would be interesting 
for a machine learning system to solve?

To avoid cluttering this news group, please reply to
my email address below and I will post a summary to 
the mailing list later. 

[1]	Kristian Lindgren, ``Evolutionary Phenomena in Simple Dynamics''
	in Artificial Life 2, pages 295-312.  Addison-Wesley, 1991.
[2]	D.Moriarty and R.Miikkulainen, ``Evolving Complex Othello
	Strategies using Marker-based Genetic Encoding of Neural 
	Networks'', Technical Report AI93-206, Dept. of Computer Science
	University of Texas at Austin, September 1993.

________________________________________________________________________

Paul Darwen                                   darwen@canth.cs.adfa.oz.au
Department of Computer Science                     Phone: +61 6 268 8182
University College, UNSW                             Fax: +61 6 268 8581
Canberra  ACT  2601  AUSTRALIA
________________________________________________________________________


