Reinforcement Learning with Function Approximation for Non-Markov Processes

Ali Devran Kara

We study reinforcement learning methods with linear function approximation under non-Markov state and cost processes. We first consider the policy evaluation method and show that t…