A Model-free Learning Algorithm for Infinite-horizon Average-reward MDPs with Near-optimal Regret
This addresses the need for efficient and flexible reinforcement learning algorithms without ergodic assumptions, though it is incremental as it builds on prior model-free and model-based work.
The paper tackles the problem of model-free reinforcement learning for infinite-horizon average-reward MDPs, proposing EE-QL which achieves a near-optimal regret bound of O(√T) for weakly communicating MDPs, matching the lower bound up to logarithmic factors and performing comparably to model-based algorithms in experiments.
Recently, model-free reinforcement learning has attracted research attention due to its simplicity, memory and computation efficiency, and the flexibility to combine with function approximation. In this paper, we propose Exploration Enhanced Q-learning (EE-QL), a model-free algorithm for infinite-horizon average-reward Markov Decision Processes (MDPs) that achieves regret bound of $O(\sqrt{T})$ for the general class of weakly communicating MDPs, where $T$ is the number of interactions. EE-QL assumes that an online concentrating approximation of the optimal average reward is available. This is the first model-free learning algorithm that achieves $O(\sqrt T)$ regret without the ergodic assumption, and matches the lower bound in terms of $T$ except for logarithmic factors. Experiments show that the proposed algorithm performs as well as the best known model-based algorithms.