Policy Invariance under Reward Transformations for General-Sum Stochastic Games
X. Lu, Howard M. Schwartz, Sidney Nascimento Givigi · Journal of Artificial Intelligence Research · 2011
We extend the potential-based shaping method from Markov decision processes to multi-player general-sum stochastic games. We prove that the Nash equilibria in a stochastic game remains unchanged after potential-based shaping is applied to the environment. The property of policy invariance provides a possible way of speeding convergence when learning to play a stochastic game.