Don't Do What Doesn't Matter: Intrinsic Motivation with Action Usefulness
This addresses the challenge of inefficient exploration in reinforcement learning for AI agents, offering a domain-specific improvement over existing intrinsic motivation methods.
The paper tackles the exploration problem in reinforcement learning with sparse rewards by proposing a new method, DoWhaM, which shifts focus from state novelty to rewarding actions that seldom affect the environment, resulting in greatly reduced sample complexity on the MiniGrid environment.
Sparse rewards are double-edged training signals in reinforcement learning: easy to design but hard to optimize. Intrinsic motivation guidances have thus been developed toward alleviating the resulting exploration problem. They usually incentivize agents to look for new states through novelty signals. Yet, such methods encourage exhaustive exploration of the state space rather than focusing on the environment's salient interaction opportunities. We propose a new exploration method, called Don't Do What Doesn't Matter (DoWhaM), shifting the emphasis from state novelty to state with relevant actions. While most actions consistently change the state when used, \textit{e.g.} moving the agent, some actions are only effective in specific states, \textit{e.g.}, \emph{opening} a door, \emph{grabbing} an object. DoWhaM detects and rewards actions that seldom affect the environment. We evaluate DoWhaM on the procedurally-generated environment MiniGrid, against state-of-the-art methods and show that DoWhaM greatly reduces sample complexity.