SOURCE-LINKED INTELLIGENCE

Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach

arXiv · AI, language, vision and robotics · article · Oct 7, 2026 · UTC

We study learning optimal policies in average-reward multichain Markov decision processes (MDPs), where the optimal gain may depend on the initial state and recurrence structures vary across policies, creating challenges for reinforcement learning (RL) methods. We propose an asynchronous value-iteration-based RL algorithm that requires no model knowledge beyond the MDP's transition graph and leverages Bather's decomposition to hierarchically partition the state space into communicating subsystems and transient states. This decomposition induces a recasting of the global decision problem into s

Read original source ↗ Open in workspace

recordType

paper

region

Global