Learning Action-Transferable Policy with Action Embedding
This addresses sample efficiency in reinforcement learning for related tasks, but it is incremental as it builds on prior work by extending state-space methods to include action spaces.
The paper tackles the problem of transferring policies across reinforcement learning tasks with different state and action spaces by proposing a method to learn action embeddings, which accelerates policy learning in experiments.
Transfer learning (TL) is a promising way to improve the sample efficiency of reinforcement learning. However, how to efficiently transfer knowledge across tasks with different state-action spaces is investigated at an early stage. Most previous studies only addressed the inconsistency across different state spaces by learning a common feature space, without considering that similar actions in different action spaces of related tasks share similar semantics. In this paper, we propose a method to learning action embeddings by leveraging this idea, and a framework that learns both state embeddings and action embeddings to transfer policy across tasks with different state and action spaces. Our experimental results on various tasks show that the proposed method can not only learn informative action embeddings but accelerate policy learning.