🚀 go-pugleaf

In article <Bsx9rB.F6H@news.cso.uiuc.edu> tmkk@uiuc.edu (Khan) writes:
>
>In article <BswJE8.261@ais.org> draper@ais.org (Patrick Draper) writes:
>>Has anyone tried using genetic algorithms for CCP's?
>>
>>I'll bet you have to play thousands of games to train a good one, unless you
>>prime the pump with abover average players first.
>>
>>Any comments?
>
>Ooh, a neural net computer opponent, which would learn from the
>programmer and beta testers while they play test the game, so by the


Actually, I was thinking about a GENETIC type learning approach. If you
define control strings for enemies, then battle and score them, you can
determine the best battlers among them. Then you randomly cross a couple,
mutate a couple, throw out the bad ones, and try them all out again.

In that way, the best ones survive to get "crossed" in the next breeding,
and the bad ones are eliminated.


------------------////////////////////\\\\\\\\\\\\\\\\\\\\------------------
| Patrick Draper           Disclaimer:  I can't control my fingers,        |
| draper@irie.ais.edu                   I can't control my toes! - Ramones |
| University of Michigan                                                   |
| Computer Club                We are a nation of laws, not people         |
------------------\\\\\\\\\\\\\\\\\\\\////////////////////------------------