@article{Kupcsik:2017:10.1016/j.artint.2014.11.005,
author = {Kupcsik, A and Deisenroth, MP and Peters, J and Loh, AP and Vadakkepat, P and Neumann, G},
doi = {10.1016/j.artint.2014.11.005},
journal = {Artificial Intelligence},
pages = {415--439},
title = {Model-based contextual policy search for data-efficient generalization of robot skills},
url = {http://dx.doi.org/10.1016/j.artint.2014.11.005},
volume = {247},
year = {2017}
}

Download

TY  - JOUR
AB  - © 2014 Elsevier B.V. In robotics, lower-level controllers are typically used to make the robot  solve a specific task in a fixed context. For example, the lower-level controller can encode a hitting movement while the context defines the target coordinates to hit. However, in many learning problems the context may change between task executions. To adapt the policy to a new context, we utilize a hierarchical approach by learning an upper-level policy that generalizes the lower-level controllers to new contexts. A common approach to learn such upper-level policies is to use policy search. However, the majority of current contextual policy search approaches are model-free and require a high number of interactions with the robot and its environment. Model-based approaches are known to significantly reduce the amount of robot experiments, however, current model-based techniques cannot be applied straightforwardly to the problem of learning contextual upper-level policies. They rely on specific parametrizations of the policy and the reward function, which are often unrealistic in the contextual policy search formulation. In this paper, we propose a novel model-based contextual policy search algorithm that is able to generalize lower-level controllers, and is data-efficient. Our approach is based on learned probabilistic forward models and information theoretic policy search. Unlike current algorithms, our method does not require any assumption on the parametrization of the policy or the reward function. We show on complex simulated robotic tasks and in a real robot experiment that the proposed learning framework speeds up the learning process by up to two orders of magnitude in comparison to existing methods, while learning high quality policies.
AU  - Kupcsik,A
AU  - Deisenroth,MP
AU  - Peters,J
AU  - Loh,AP
AU  - Vadakkepat,P
AU  - Neumann,G
DO  - 10.1016/j.artint.2014.11.005
EP  - 439
PY  - 2017///
SN  - 0004-3702
SP  - 415
TI  - Model-based contextual policy search for data-efficient generalization of robot skills
T2  - Artificial Intelligence
UR  - http://dx.doi.org/10.1016/j.artint.2014.11.005
UR  - http://hdl.handle.net/10044/1/19434
VL  - 247
ER  - 

Download