Timezone: »

Learning to Execute: Efficient Learning of Universal Plan-Conditioned Policies in Robotics
Ingmar Schubert · Danny Driess · Ozgur S. Oguz · Marc Toussaint

Wed Dec 08 12:30 AM -- 02:00 AM (PST) @

Applications of Reinforcement Learning (RL) in robotics are often limited by high data demand. On the other hand, approximate models are readily available in many robotics scenarios, making model-based approaches like planning a data-efficient alternative. Still, the performance of these methods suffers if the model is imprecise or wrong. In this sense, the respective strengths and weaknesses of RL and model-based planners are complementary. In the present work, we investigate how both approaches can be integrated into one framework that combines their strengths. We introduce Learning to Execute (L2E), which leverages information contained in approximate plans to learn universal policies that are conditioned on plans. In our robotic manipulation experiments, L2E exhibits increased performance when compared to pure RL, pure planning, or baseline methods combining learning and planning.

Author Information

Ingmar Schubert (Technische Universität Berlin / Learning and Intelligent Systems Group)
Danny Driess (TU Berlin)
Ozgur S. Oguz (Uni Stuttgart & MPI - IS)
Marc Toussaint (Universty Stuttgart)

More from the Same Authors