Inverse Optimal Control

In Reinforcement Learning, an agent learns a policy that maximizes a given reward function. However, providing a reward function for a given learning task is often non trivial. Inverse Reinforcement Learning, which is sometimes also called Inverse Optimal Control, addresses this problem by learning the reward function from expert demonstrations. The aim of this paper is to give a brief introduction to Inverse Reinforcement Learning. It thereby focuses on two recent approaches which employ Gaussian Processes to learn nonlinear reward functions and maximum causal entropy to model the experts behavior. While the first approach is infeasible if the state space is too large, the second, more recent approach was successfully applied to high dimensional and continuous learning tasks.