Subwords as Skills: Tokenization for Sparse-Reward Reinforcement Learning
This addresses the problem of efficient exploration in sparse-reward RL for continuous action spaces, offering a more computationally efficient alternative to existing skill-generation methods.
The paper tackles the challenge of exploration in sparse-reward reinforcement learning by proposing a novel skill-generation method that discretizes the action space and uses tokenization to create temporally extended actions, resulting in outperforming baselines and requiring orders-of-magnitude less computation.
Exploration in sparse-reward reinforcement learning is difficult due to the requirement of long, coordinated sequences of actions in order to achieve any reward. Moreover, in continuous action spaces there are an infinite number of possible actions, which only increases the difficulty of exploration. One class of methods designed to address these issues forms temporally extended actions, often called skills, from interaction data collected in the same domain, and optimizes a policy on top of this new action space. Typically such methods require a lengthy pretraining phase, especially in continuous action spaces, in order to form the skills before reinforcement learning can begin. Given prior evidence that the full range of the continuous action space is not required in such tasks, we propose a novel approach to skill-generation with two components. First we discretize the action space through clustering, and second we leverage a tokenization technique borrowed from natural language processing to generate temporally extended actions. Such a method outperforms baselines for skill-generation in several challenging sparse-reward domains, and requires orders-of-magnitude less computation in skill-generation and online rollouts. Our code is available at \url{https://github.com/dyunis/subwords_as_skills}.