Versatile Skill Control via Self-supervised Adversarial Imitation of Unlabeled Mixed Motions
Sep 15, 2022
Reinforcement Learning for Robotics
My research interests focus on the general field of robot learning, including reinforcement learning, developmental robotics and legged intelligence.
In this work, we propose a cooperative generative adversarial method for obtaining controllable skill sets from unlabeled datasets containing diverse state transition patterns.
Chenhao Li, Sebastian Blaes, Pavel Kolev, Marin Vlastelica, Jonas Frey, Georg Martius
Works that propose possible solutions to reduce such human involvement, specifically with examples of legged skill development using model-free reinforcement learning and imitation learning methods.
Jan 12, 2023 4:00 PM Biomimetic Robotics Lab, Massachusetts Institute of Technology
Methods that allow skill transfer from limited and noisy reference demonstrations.
Dec 2, 2022 5:00 PM Machines in Motion Laboratory, New York University