Demonstration-Bootstrapped Autonomous Practicing via Multi-Task Reinforcement Learning
Demonstration-Bootstrapped Autonomous Practicing via Multi-Task Reinforcement Learning
Reinforcement learning systems have the potential to enable continuous improvement in unstructured environments, leveraging data collected autonomously. However, in practice these systems require significant amounts of instrumentation or human intervention to learn in the real world. In this work, we propose a system for reinforcement learning that leverages multi-task reinforcement …