Action Redundancy in Reinforcement Learning

TL;DR

This paper introduces the concept of transition entropy in reinforcement learning, focusing on reducing action redundancy to improve learning efficiency, with algorithms tested on synthetic and benchmark environments.

Contribution

It proposes a novel focus on transition entropy and develops algorithms to minimize action redundancy, addressing a fundamental problem in reinforcement learning.

Findings

Action redundancy significantly impacts RL performance.

Algorithms effectively reduce redundancy in synthetic and benchmark tests.

Transition entropy offers a new perspective beyond action entropy.

Abstract

Maximum Entropy (MaxEnt) reinforcement learning is a powerful learning paradigm which seeks to maximize return under entropy regularization. However, action entropy does not necessarily coincide with state entropy, e.g., when multiple actions produce the same transition. Instead, we propose to maximize the transition entropy, i.e., the entropy of next states. We show that transition entropy can be described by two terms; namely, model-dependent transition entropy and action redundancy. Particularly, we explore the latter in both deterministic and stochastic settings and develop tractable approximation methods in a near model-free setup. We construct algorithms to minimize action redundancy and demonstrate their effectiveness on a synthetic environment with multiple redundant actions as well as contemporary benchmarks in Atari and Mujoco. Our results suggest that action redundancy is a fundamental problem in reinforcement learning.