Subject: Re: Squad tactics/ was Re: What's going wrong with this newsgroup?
References: <4i6lav$t5m@maze.ruca.ua.ac.be> <31478DC7.3F7F@ultranet.com> <3149C3B7.3811@3rdplanet.com> <4iqm3c$ptp@tribune.cris.com> <4isp30$qgb@elaine46.Stanford.EDU>
X-Newsreader: TIN [version 1.2 PL2]
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!scramble.lm.com!news.math.psu.edu!psuvax1!news.cc.swarthmore.edu!netnews.upenn.edu!msunews!uwm.edu!cs.utexas.edu!howland.reston.ans.net!tank.news.pipex.net!pipex!lade.news.pipex.net!pipex!news00.sunet.se!sunic!news99.sunet.se!news.uni-c.dk!imada.ou.dk!breese
From: breese@imada.ou.dk (Bjorn Reese)
Message-ID: <1996Mar23.140556.23666@imada.ou.dk>
Sender: news@imada.ou.dk
Nntp-Posting-Host: wagner.imada.ou.dk
Organization: Dept. of Math. & Computer Science, Odense University, Denmark
Date: Sat, 23 Mar 1996 14:05:56 GMT
Lines: 22
Newsgroups: comp.ai.games

Matthew David Hills (hills@leland.Stanford.EDU) wrote:
> So, what is the advantage of generating a computer player by genetic
> algorithms over a more traditional search method?

Genetic algorithms make random picks. If a good solution is found
the search (the random picking) is intensified in the "neighbourhood"
of that solution. Genetic algorithms can quickly find good solutions
in complex search domains - as opposed to exhausive seach - but are
never guaranteed to find the best solution. If your search domain is
simple or there already exits another search method for that domain,
GAs are likely to loose.

Slightly off the topic, I think GA is too hyped. GA is only one
method in the field of Evolutionary Computation. My personal favorite
is Evolutionary Programming, which works pretty much like GA, but you
don't have to force your data into bit-strings.

--
Bjorn Reese                      Email: breese@imada.ou.dk
Odense University, Denmark       URL:   http://www.imada.ou.dk/~breese

"It's getting late in the game to show any pride or shame" - Marillion
