Dr. Sergey Levine
VERIFIED TECHNICAL DOSSIERSources checked

Dr. Sergey Levine

UC Berkeley / Physical Intelligence

2 Verified ArtifactsSource Checked & Attributed

Verified Proof of Work Artifacts

2 items cataloged

Each artifact below represents an authenticated research publication, production code repository, or technical architectural framework directly authored or co-created by Dr. Sergey Levine. Every entry undergoes editorial source verification.

#1
RESEARCH Checked Sep 21, 2026

End-to-End Training of Deep Visuomotor Policies

Published the foundational Journal of Machine Learning Research paper proving that robotic controllers can map raw camera pixels directly to continuous motor torque voltages using guided policy search.

Model & Execution Context:Convolutional neural network visuomotor policies operating on physical PR2 robotic systems.
Scope & Limitations

Direct pixel-to-torque policies are vulnerable to novel environmental lighting and visual background shifts.

#2
RESEARCH Checked Sep 21, 2026

π₀ (pi_0): Generalist Physical Intelligence Robot Foundation Model

Co-developed π₀, a generalist physical intelligence foundation model trained across diverse robotic embodiments and tasks, demonstrating zero-shot physical manipulation, folding laundry, and assembly.

Model & Execution Context:Flow-matching vision-action transformer architecture controlling multi-degree-of-freedom robotic arms.
Scope & Limitations

Physical real-world manipulation requires sub-millisecond control loops and collision avoidance safeguards.