Accelerating Representation Learning with View-Consistent Dynamics in Data-Efficient Reinforcement Learning - 42Papers