Hi Guys , You've got a great forum here, I was researching in Adaptation ie "Planning and Learning" in RTS-Games I figured out there many approaches for this such as : 1)Case-Based Planning 2)Reinforcement Learning a)Dynamic Scripting b)Monte-Carlo Methods c)Q-learning d)TD-learning 3)Artificial Neural Networks 4)Genetic Algorithms 5)Hybrid CBR/Reinforcement Learning I read about all of these, I've got papers in all of them, I'd like to get your opinion or your experience, I want to focus research on one of them so i want to know from your points of views which ones are the most promising techniques(which need much research to give better results) please tell me if i forgot something, or misunderstood another ! thxs for precious time, Omar

For what it's worth we discussed this already on The Game AI Forums, and I just added a new reply to add my opinions on the subject.

http://forums.aigamedev.com/showthread.php?t=2952

thxs for answering there Alexjc :)

and thxs for telling me , cuz i forgot totally about the thread in the other forum.