Mar 5, 2019 · We introduce Play-LMP, a self-supervised method that learns to organize play behaviors in a latent space, then reuse them at test time to achieve specific ...

Mar 5, 2019 · We propose learning from teleoperated play data as a way to scale up multi-task robotic skill learning. Learning from play (LfP) offers ...

(a) Training: 1) sample a random window of experience from a memory of play data; 2) train to recognize and organize a repertoire of behaviors executed during ...

We propose a self-supervised approach to learning a wide variety of manipulation skills from unlabeled data collected through playing in and interacting ...

Oct 11, 2022 · ❖ Play-supervised Latent Motor Plans: learning representations of all the different high-level plans ( p(b|sc,sg) )and condition a policy ...

We propose learning from teleoperated play data as a way to scale up multi-task robotic skill learning. Learning from play (LfP) offers three main ...

Mar 5, 2019 · We propose learning from teleoperated play data (LfP) as a way to scale up multi-task robotic skill learning. Learning from play (LfP) ...

This work proposes to use playful interactions in a self-supervised manner to learn visual representations for downstream tasks.

In this work, we proposed learning many skills at once from unstructured human "play" data as a way to scale up multitask learning. Here, a human tele-operates ...

Mar 5, 2019 · Abstract—We propose learning from teleoperated play data. (LfP) as a way to scale up multi-task robotic skill learning.