Stanford Seminar - Robot Learning in the Era of Large Pretrained Models

February 23, 2024
Dorsa Sadigh, Stanford University

In this talk, I will discuss how interactive robot learning can benefit from the rise of large pretrained models such as foundation models. I will introduce two perspectives. First I will discuss the role of pretraining when learning visual representations, and how language can guide learning grounded visual representations useful for downstream robotics tasks. I will then discuss the choice of datasets during pretraining. Specifically, how we could guide large scale data collection, and what constitutes high quality data for imitation learning. I will discuss some recent work around guiding data collection based on enabling compositional generalization of learned policies. Finally, I will end the talk by discussing a few creative ways of tapping into the rich context of large language models and vision-language models for robotics.

[cardianlfan]

The Age of Robots has arrived. With the unified interface for large-model robot hardware, we can undoubtedly achieve the same astonishing results as GPT language systems.

[cedricmanouan]

great talk!
The discussion on guided data collection is pretty informative

[jaiberjohn]

It'd be good to listen to the skipped content, including reinforcement learning, and more details on the Take 2 items. Great presentation!

[TV]

robot learning is 76% understanding education level i assume so
artificial intelligence use
study education quickly 😅