Exploring Unknown States with Action Balance

TL;DR

This paper introduces action balance exploration, an extension of UCB for deep reinforcement learning, which improves the discovery of unknown states and enhances exploration efficiency in challenging environments.

Contribution

It proposes a novel action balance exploration method and combines it with RND to better explore unknown states in reinforcement learning environments.

Findings

Action balance exploration outperforms traditional methods in finding unknown states.

Combining action balance with RND improves performance in hard exploration environments.

Experiments on grid world and Atari demonstrate enhanced exploration capabilities.

Abstract

Exploration is a key problem in reinforcement learning. Recently bonus-based methods have achieved considerable successes in environments where exploration is difficult such as Montezuma's Revenge, which assign additional bonuses (e.g., intrinsic rewards) to guide the agent to rarely visited states. Since the bonus is calculated according to the novelty of the next state after performing an action, we call such methods as the next-state bonus methods. However, the next-state bonus methods force the agent to pay overmuch attention in exploring known states and ignore finding unknown states since the exploration is driven by the next state already visited, which may slow the pace of finding reward in some environments. In this paper, we focus on improving the effectiveness of finding unknown states and propose action balance exploration, which balances the frequency of selecting each action at a given state and can be treated as an extension of upper confidence bound (UCB) to deep reinforcement learning. Moreover, we propose action balance RND that combines the next-state bonus methods (e.g., random network distillation exploration, RND) and our action balance exploration to take advantage of both sides. The experiments on the grid world and Atari games demonstrate action balance exploration has a better capability in finding unknown states and can improve the performance of RND in some hard exploration environments respectively.