199259
20180913062523.0
978-1-4673-6358-7
2153-0858
000331367403046
ISI
CONF
Transfer in Inverse Reinforcement Learning for Multiple Strategies
New York
2013
Ieee
2013
7
Conference Papers
IEEE International Conference on Intelligent Robots and Systems
We consider the problem of incrementally learning different strategies of performing a complex sequential task from multiple demonstrations of an expert or a set of experts. While the task is the same, each expert differs in his/her way of performing it. We assume that this variety across experts' demonstration is due to the fact that each expert/strategy is driven by a different reward function, where reward function is expressed as a linear combination of a set of known features. Consequently, we can learn all the expert strategies by forming a convex set of optimal deterministic policies, from which one can match any unseen expert strategy drawn from this set. Instead of learning from scratch every optimal policy in this set, the learner transfers knowledge from the set of learned policies to bootstrap its search for new optimal policy. We demonstrate our approach on a simulated mini-golf task where the 7 degrees of freedom Barrett WAM robot arm learns to sequentially putt on different holes in accordance with the playing strategies of the expert.
Tanwani, Ajay Kumar
216104
246728
Billard, Aude
115671
240594
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Amato, N.
ed.
3244-3250
2013 Ieee/Rsj International Conference On Intelligent Robots And Systems (Iros)
LASA
252119
U10660
oai:infoscience.tind.io:199259
STI
conf
115671
EPFL-CONF-199259
EPFL
PUBLISHED
REVIEWED
CONF