Research topic
Background: Generative policies and multimodal actions. 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 ยท ~37 minutes
Explore the labRecommended through the concepts linked to this topic. Each lesson belongs to an existing learning series.
How Robots Learn
Generative models for robot controlExplain why averaging distinct valid actions can be unsafe.
Supports: Generative policies
2025
Park S, Li Q, Levine S
Why this matters: Berkeley / RAIL research. Combines a generative policy with value-guided learning.