A Generalist Dynamics Model for Control

no code implementations • 18 May 2023 • Ingmar Schubert, Jingwei Zhang, Jake Bruce, Sarah Bechtle, Emilio Parisotto, Martin Riedmiller, Jost Tobias Springenberg, Arunkumar Byravan, Leonard Hasenclever, Nicolas Heess

We investigate the use of transformer sequence models as dynamics models (TDMs) for control.

Reinforcement Learning with Neural Radiance Fields

no code implementations • 3 Jun 2022 • Danny Driess, Ingmar Schubert, Pete Florence, Yunzhu Li, Marc Toussaint

This paper demonstrates that learning state representations with supervision from Neural Radiance Fields (NeRFs) can improve the performance of RL compared to other learned representations or even low-dimensional, hand-engineered state information.

reinforcement-learning Reinforcement Learning (RL)

Learning to Execute: Efficient Learning of Universal Plan-Conditioned Policies in Robotics

1 code implementation • NeurIPS 2021 • Ingmar Schubert, Danny Driess, Ozgur S. Oguz, Marc Toussaint

Applications of Reinforcement Learning (RL) in robotics are often limited by high data demand.

Learning to Execute Reinforcement Learning (RL)

Plan-Based Relaxed Reward Shaping for Goal-Directed Tasks

no code implementations • 14 Jul 2021 • Ingmar Schubert, Ozgur S. Oguz, Marc Toussaint

In high-dimensional state spaces, the usefulness of Reinforcement Learning (RL) is limited by the problem of exploration.

Reinforcement Learning (RL)

Plan-Based Asymptotically Equivalent Reward Shaping

no code implementations • ICLR 2021 • Ingmar Schubert, Ozgur S Oguz, Marc Toussaint

In high-dimensional state spaces, the usefulness of Reinforcement Learning (RL) is limited by the problem of exploration.

reinforcement-learning Reinforcement Learning (RL)