Experts
EN·Also available in French — switch via the flag
P
Expert
Pieter Abbeel
Robot LearningReinforcement LearningRobotic ManipulationWarehouse RoboticsFoundation Models for Robotics
About
Pieter Abbeel is a professor at UC Berkeley, director of the Berkeley Robot Learning Lab and co-founder of Covariant. His research includes reinforcement learning, robot learning and manipulation.
His work advances robot learning and generalizable robotic intelligence.
Dataleo perspective
Learning robots need governed training data, simulation-to-reality testing, confidence thresholds and safe fallback behavior before autonomous deployment.
Teams should document model versions, item coverage and intervention rules for reinforcement learning systems.
Related content