CaT: Constraints as Terminations for Legged Locomotion Reinforcement Learning

no code implementations • 27 Mar 2024 • Elliot Chane-Sane, Pierre-Alexandre Leziart, Thomas Flayols, Olivier Stasse, Philippe Souères, Nicolas Mansard

In this work, we advocate for integrating constraints into robot learning and present Constraints as Terminations (CaT), a novel constrained RL algorithm.

reinforcement-learning Reinforcement Learning (RL)

Learning Video-Conditioned Policies for Unseen Manipulation Tasks

no code implementations • 10 May 2023 • Elliot Chane-Sane, Cordelia Schmid, Ivan Laptev

To encourage generalization to new tasks, we avoid particular tasks during training and learn our policy from unlabelled robot trajectories and corresponding robot videos.

Action Recognition Robot Manipulation +1

Goal-Conditioned Reinforcement Learning with Imagined Subgoals

no code implementations • 1 Jul 2021 • Elliot Chane-Sane, Cordelia Schmid, Ivan Laptev

Goal-conditioned reinforcement learning endows an agent with a large variety of skills, but it often struggles to solve tasks that require more temporally extended reasoning.

reinforcement-learning Reinforcement Learning (RL)