Bellman Optimality of Average-Reward Robust Markov Decision Processes with a Constant Gain
This work develops a framework for finding optimal control strategies in robust Markov Decision Processes (MDPs) when the goal is to maximize long-run average reward. By analyzing the constant-gain setting, it establishes conditions under which solutions to the robust Bellman equation exist and characterize the optimal average reward and stationary policies, even when there is information asymmetry with an adversary.