Research topic
Background: Cross-embodiment training and hierarchical control. Evaluation limits and transfer uncertainty remain explicit.
Independent educational guide. Not affiliated with or endorsed by the universities, professors or laboratories described here. This collection reflects the material currently mapped on Socratic Learn.
Berkeley RAIL research studies robot learning from demonstrations, reward, prior datasets, and interaction: policies, reinforcement learning, manipulation, and generalization. The independent Levine primer also covers his co-founder work at Physical Intelligence, a separate company.
7 lessons · ~37 minutes
Explore the labRecommended through the concepts linked to this topic. Each lesson belongs to an existing learning series.
How Robots Learn
Physical Intelligence: building a foundation model for robotsDistinguish a VLM from a VLA and a company goal from demonstrated capability.
Supports: Cross-embodiment learning
2023
Shah D, Sridhar A, Dashora N, Stachowicz K, Black K, Hirose N, Levine S
Why this matters: Berkeley / RAIL research. A visual-navigation foundation model evaluated across robot platforms.
2026
Physical Intelligence
Why this matters: Physical Intelligence current research: context-steered transfer examples with lower reported success on unseen tasks; transfer remains unresolved.