Zero-Shot Skill Composition and Simulation-to-Real Transfer by Learning Task Representations

Zhanpeng He
Zhanpeng He
Ryan Julian
Ryan Julian
Eric Heiden
Eric Heiden
Hejia Zhang
Hejia Zhang

arXiv: Robotics, Volume abs/1810.02422, 2018.

Cited by: 5|Bibtex|Views35|Links
EI

Abstract:

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

Code:

Data:

Your rating :
0

 

Tags
Comments