genetic algorithm to optimize static evaluation function

Hi guys. I'm supposed to do a computer game for a school project. So I chose five in a row. I'm using a standard minimax algorithm. I also want to somehow optimize the static evaluation function, so I thought about using genetic algorithm to do this. The evaluation function has some parameters, so I would generate several sets of those parametrs and let them play against each other, and then I'd replace the one with the poorest result with some combination of the two with the best result and this would go on for some time. Question is: Could this produce a viable result or is it inapropriate to use genetic algorithm for this problem? I mean doesn't it create some problems because the result of one set depends on the other ones?
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!