🚀 go-pugleaf

In article <1992Aug16.071200.10997@pasteur.Berkeley.EDU> pychen@cory.Berkeley.EDU (Peter Y. Chen) writes:
>In article <BswJE8.261@ais.org> draper@ais.org (Patrick Draper) writes:
>>Has anyone tried using genetic algorithms for CCP's?
>		        ^^^^^^^^^^^^^^^^^^
>>
>
>Is this like simulated evolution algorithm?  Please enlighten, I would
>like to know more about this.
>
I think you maybe talking about the same term here, although almost everyone 
tends to use the term 'genetic algorithm (GA)'.  

Genetic algorithms are based on the reproductive algorithms in nature,
and hence it is possible to perform simulated evolution using the GA.
However, simulated evolution usually never just consists of GAs.  Neural 
networks etc are also important.

Back to the original question, I guess it would be possible to use GAs
for CCPs.  The way to do this (IMHO) would be to define a set of
attributes for each CCP and then create a chromosomal representation for
the list of attributes.  If you have a scenario where the CCPs must
survive then work out an evaluation function to calculate which CCPs are
the strongest and hence which move onto the next round in the game.  
It would be quite simple to create new CCPs by crossing over (and
mutating) the attributes of other CCPs.  

So a simple algorithm could be:
1. Create a random population of CCPs with various attributes.
2. Denote a fitness value to each CCP by using some sort of evaluation
   function.
3. Do something with the CCPs
4. Create new CCPs by mating current CCPs.
5. Delete weak CCPs and replace them with the new CCPs.

>-- Peter
>

Ross.

-- 
Ross Milward,   ross@coral.cs.jcu.edu.au, cprem@marlin.jcu.edu.au
Honours Student - Comp. Sci. Dept.,  James Cook Uni, Qld, 4811, Australia.
"The way the rain comes down hard, that's how I feel inside." - The Cure