Where to Intervene: Action Selection in Deep Reinforcement Learning

TL;DR

This paper introduces a data-driven, model-free action selection method for deep reinforcement learning that improves efficiency and generalizability by selecting minimal sufficient actions and controlling false discoveries, enhancing performance across tasks.

Contribution

It proposes a novel, computationally efficient action selection approach that integrates into deep RL, addressing high-dimensional action spaces without relying on domain-specific priors.

Findings

Outperforms existing methods in variable selection accuracy

Achieves higher overall rewards in empirical tests

Provides theoretical guarantees for false discovery control

Abstract

Deep reinforcement learning (RL) has gained widespread adoption in recent years but faces significant challenges, particularly in unknown and complex environments. Among these, high-dimensional action selection stands out as a critical problem. Existing works often require a sophisticated prior design to eliminate redundancy in the action space, relying heavily on domain expert experience or involving high computational complexity, which limits their generalizability across different RL tasks. In this paper, we address these challenges by proposing a general data-driven action selection approach with model-free and computationally friendly properties. Our method not only selects minimal sufficient actions but also controls the false discovery rate via knockoff sampling. More importantly, we seamlessly integrate the action selection into deep RL methods during online training. Empirical experiments validate the established theoretical guarantees, demonstrating that our method surpasses various alternative techniques in terms of both performance in variable selection and overall achieved rewards.