A Physics-Based Model Prior for Object-Oriented MDPs


Jonathan Scholz, Martin Levihn, Charles Isbell, David Wingate ;
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):1089-1097, 2014.


One of the key challenges in using reinforcement learning in robotics is the need for models that capture natural world structure. There are, methods that formalize multi-object dynamics using relational representations, but these methods are not sufficiently compact for real-world robotics. We present a physics-based approach that exploits modern simulation tools to efficiently parameterize physical dynamics. Our results show that this representation can result in much faster learning, by virtue of its strong but appropriate inductive bias in physical environments.

Related Material