interpretable dynamics models for data-efficient reinforcement learning

Markus Kaiser, Clemens Otte, Thomas A. Runkler, Carl Henrik Ek