Demonstration-Guided Reinforcement Learning with Learned Skills
Pertsch, Karl, Lee, Youngwoon, Wu, Yue, Lim, Joseph J.
–arXiv.org Artificial Intelligence
Demonstration-guided reinforcement learning (RL) is a promising approach for learning complex behaviors by leveraging both reward feedback and a set of target task demonstrations. Prior approaches for demonstration-guided RL treat every new task as an independent learning problem and attempt to follow the provided demonstrations step-by-step, akin to a human trying to imitate a completely unseen behavior by following the demonstrator's exact muscle movements. Naturally, such learning will be slow, but often new behaviors are not completely unseen: they share subtasks with behaviors we have previously learned. In this work, we aim to exploit this shared subtask structure to increase the efficiency of demonstration-guided RL. We first learn a set of reusable skills from large offline datasets of prior experience collected across many tasks. We then propose Skill-based Learning with Demonstrations (SkiLD), an algorithm for demonstration-guided RL that efficiently leverages the provided demonstrations by following the demonstrated skills instead of the primitive actions, resulting in substantial performance improvements over prior demonstration-guided RL approaches. We validate the effectiveness of our approach on long-horizon maze navigation and complex robot manipulation tasks.
arXiv.org Artificial Intelligence
Jul-21-2021
- Country:
- North America > United States > California (0.14)
- Genre:
- Research Report > Promising Solution (0.48)
- Industry:
- Education > Focused Education (0.34)
- Technology: