Leonel Dario Rozo

Papers

Total Citations

H-Index

About

Leonel Dario Rozo is a leading researcher at the intersection of robotics and machine learning, with a core focus on developing data-efficient, geometry-aware algorithms for robot learning. His major contributions lie in advancing Bayesian optimization (BO) for high-dimensional and non-Euclidean parameter spaces, a critical challenge in robotics. Rozo pioneered the integration of Riemannian geometry with BO, introducing Riemannian Matérn kernels that enable effective optimization on manifolds like spheres and rotation groups—domains fundamental to robotic control and perception. His work on geometry-aware BO, detailed in papers with 10 and 5 citations respectively, directly addresses the limitations of standard BO in robotics, offering a principled framework for tuning control parameters and adapting policies with minimal data. Additionally, Rozo developed PyRoboLearn (12 citations), a Python framework designed to democratize robot learning for practitioners. By bridging theoretical advances in Bayesian optimization with practical tools, Rozo has significantly enhanced the efficiency and applicability of robot learning, making his research indispensable for students and engineers seeking to optimize robotic systems in complex, real-world environments.