Control Frequency Adaptation via Action Persistence in Batch Reinforcement Learning
This addresses a domain-specific issue in reinforcement learning for improving policy performance through control frequency adaptation, representing an incremental advancement.
The paper tackles the problem of selecting an optimal control frequency in reinforcement learning by introducing action persistence, which repeats actions over multiple steps, and presents Persistent Fitted Q-Iteration (PFQI) to learn optimal policies at a given persistence, with experiments on benchmark domains showing effectiveness.
The choice of the control frequency of a system has a relevant impact on the ability of reinforcement learning algorithms to learn a highly performing policy. In this paper, we introduce the notion of action persistence that consists in the repetition of an action for a fixed number of decision steps, having the effect of modifying the control frequency. We start analyzing how action persistence affects the performance of the optimal policy, and then we present a novel algorithm, Persistent Fitted Q-Iteration (PFQI), that extends FQI, with the goal of learning the optimal value function at a given persistence. After having provided a theoretical study of PFQI and a heuristic approach to identify the optimal persistence, we present an experimental campaign on benchmark domains to show the advantages of action persistence and proving the effectiveness of our persistence selection method.