Online Reinforcement Learning in Periodic MDP

no code implementations • 16 Mar 2023 • Ayush Aniket, Arpan Chattopadhyay

We study learning in periodic Markov Decision Process (MDP), a special type of non-stationary MDP where both the state transition probabilities and reward functions vary periodically, under the average reward maximization setting.

reinforcement-learning Reinforcement Learning (RL)

Online Reinforcement Learning for Periodic MDP

no code implementations • 25 Jul 2022 • Ayush Aniket, Arpan Chattopadhyay

We study learning in periodic Markov Decision Process(MDP), a special type of non-stationary MDP where both the state transition probabilities and reward functions vary periodically, under the average reward maximization setting.

reinforcement-learning Reinforcement Learning (RL)