no code implementations • 1 Jun 2018 • Ramtin Keramati, Jay Whang, Patrick Cho, Emma Brunskill
People seem to build simple models that are easy to learn to support planning and strategic exploration.
Model-based Reinforcement Learning Object +2