Scalable Option Learning in High-Throughput Environments
This work addresses the scalability problem in hierarchical reinforcement learning for researchers and practitioners, though it is incremental as it builds on existing hierarchical RL approaches.
The paper tackled the challenge of scaling hierarchical reinforcement learning to high-throughput environments by proposing the Scalable Option Learning (SOL) algorithm, which achieved a ~35x higher throughput compared to existing methods and was trained on 30 billion frames in NetHack, surpassing flat agents.
Hierarchical reinforcement learning (RL) has the potential to enable effective decision-making over long timescales. Existing approaches, while promising, have yet to realize the benefits of large-scale training. In this work, we identify and solve several key challenges in scaling online hierarchical RL to high-throughput environments. We propose Scalable Option Learning (SOL), a highly scalable hierarchical RL algorithm which achieves a ~35x higher throughput compared to existing hierarchical methods. To demonstrate SOL's performance and scalability, we train hierarchical agents using 30 billion frames of experience on the complex game of NetHack, significantly surpassing flat agents and demonstrating positive scaling trends. We also validate SOL on MiniHack and Mujoco environments, showcasing its general applicability. Our code is open sourced at: github.com/facebookresearch/sol.