Multiagent Q-Learning : Preliminary Study on Dominance between the Nash and Stackelberg Equilibriums