Keywords: Model-based Reinforcement Learning, Planning, Robotics, Model Predictive Control, Learning
Abstract: There is a widespread intuition that model-based control methods should be able to surpass the data efficiency of model-free approaches. In this paper we attempt to evaluate this intuition on various challenging locomotion tasks. We take a hybrid approach, combining model predictive control (MPC) with a learned model and model-free policy learning; the learned policy serves as a proposal for MPC. We show that MPC with learned proposals and models (trained on the fly or transferred from related tasks) can significantly improve performance and data efficiency with respect to model-free methods. However, we find that well-tuned model-free agents are strong baselines even for high DoF control problems. Finally, we show that it is possible to distil a model-based planner into a policy that amortizes the planning computation without any loss of performance.
One-sentence Summary: We combine MPC with model-free RL and evaluate on continuous control tasks from scratch and in transfer settings; our results show that model-free RL is a strong baseline in single task settings and model-based methods shine in multi-goal tasks.
Supplementary Material: zip