We introduce the use of learned shaping rewards in reinforcement learning tasks, where an agent uses prior experience on a sequence of tasks to learn a portable predictor that est...
Lyapunov design methods are used widely in control engineering to design controllers that achieve qualitative objectives, such as stabilizing a system or maintaining a system'...
This paper analyzes the complexity of on-line reinforcement learning algorithms, namely asynchronous realtime versions of Q-learning and value-iteration, applied to the problem of...
An Unobservable MDP (UMDP) is a POMDP in which there are no observations. An Only-Costly-Observable MDP (OCOMDP) is a POMDP which extends an UMDP by allowing a particular costly a...
Learning on real robots in an real, unaltered environment provides an extremely challenging problem. Many of the simplifying assumptions made in other areas of learning cannot be ...