Tony Jebara

Papers

Total Citations

256

H-Index

About

Tony Jebara is a leading researcher whose work spans machine learning, robotics, and computer vision, with a particular focus on enabling machines to learn from and interact with the physical world. His most influential contribution is the development of an SVM learning approach to robotic grasping, a seminal paper with over 215 citations that addresses the high-dimensional challenge of determining stable grasps for arbitrary objects. This work has been foundational for the field of robotic manipulation, providing a principled statistical framework for a problem that had long resisted robust solutions. Jebara has also made significant contributions to surgical robotics, co-authoring a novel drill set for enhancing and assessing robotic surgical performance, which has practical implications for medical training. His research on statistical imitative learning from perceptual data has advanced the understanding of how robots can acquire complex behaviors by observing humans, offering a more accessible alternative to traditional supervised or reinforcement learning. Through his work on modularity and specialized learning, Jebara has further shaped the theoretical foundations of behavior-based AI, arguing for focused, task-specific adaptation in resource-constrained agents. His career reflects a sustained commitment to bridging statistical learning theory with real-world robotic applications.