Zero-Shot Skill Composition and Simulation-to-Real Transfer by Learning Task Representations
arXiv: Robotics, Volume abs/1810.02422, 2018.
Simulation-to-real transfer is an important strategy for making reinforcement learning practical with real robots. Successful sim-to-real transfer systems have difficulty producing policies which generalize across tasks, despite training for thousands of hours equivalent real robot time. To address this shortcoming, we present a novel app...More
Full Text (Upload PDF)
PPT (Upload PPT)