Efficient Exploration through Intrinsic Motivation Learning for Unsupervised Subgoal Discovery in Model-Free Hierarchical Reinforcement Learning
This addresses the problem of exploration bottlenecks in HRL for researchers, but it appears incremental as it builds on existing intrinsic motivation and subgoal discovery techniques.
The paper tackled the challenge of efficient exploration for automatic subgoal discovery in Hierarchical Reinforcement Learning by showing that intrinsic motivation learning increases exploration efficiency, leading to successful subgoal discovery, with a model-free method based on unsupervised learning over a limited memory of agent experiences.
Efficient exploration for automatic subgoal discovery is a challenging problem in Hierarchical Reinforcement Learning (HRL). In this paper, we show that intrinsic motivation learning increases the efficiency of exploration, leading to successful subgoal discovery. We introduce a model-free subgoal discovery method based on unsupervised learning over a limited memory of agent's experiences during intrinsic motivation. Additionally, we offer a unified approach to learning representations in model-free HRL.