Brijen Thananjeyan
University of California, Berkeley, Baton Rouge Clinic, Santa Clara University
Papers
35
Total Citations
868
H-Index
16
About
Brijen Thananjeyan is a robotics and machine learning researcher whose work sits at the intersection of robot learning, deformable object manipulation, and surgical automation. He has made significant contributions to imitation learning and reinforcement learning for complex robotic tasks, developing algorithms such as SWIRL—a sequential inverse reinforcement learning approach for delayed-reward settings—and SAVED, a safe model-based reinforcement learning framework that leverages demonstrations to handle sparse costs and dynamical uncertainty in robotics. His research on deformable object manipulation, including fabric smoothing and rope manipulation using deep imitation learning and dense visual correspondences, has garnered substantial attention, with key papers accumulating over 100 citations each. Thananjeyan has also advanced surgical robotics, demonstrating that deep learning-based calibration can enable robotic systems to exceed human speed and consistency in surgical peg transfer tasks, and developing frameworks for high-precision surgical manipulation robust to instrument changes. His work on LazyDAgger addresses the practical burden of human supervision during interactive robot learning. Collectively, his publications have earned over 590 citations, reflecting their broad influence across robot learning, healthcare automation, and manipulation research communities.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9Learning to Localize, Grasp, and Hand Over Unmodified Surgical Needles30 citations · 2022
- 10LazyDAgger: Reducing Context Switching in Interactive Imitation Learning29 citations · 2021