Pretraining Policies
Chelsea Finn discusses the progress and challenges of pretraining policies in robot learning. While there have been successful experiments, there is still a need to make robot learning easier and more efficient. The infrastructure and setup required for robot experiments can be time-consuming and challenging, making it a research effort to set up new tasks for robots.In this clip
From this podcast
The Gradient
Chelsea Finn on Meta Learning & Model Based Reinforcement Learning
Related Questions