Goitom Leaku
Goitom Leaku
His PhD research focuses on Split Learning across underwater and surface devices within the ANT MSCA DN, aiming at efficient machine learning frameworks for resource-constrained underwater robotics operating under energy and communication limitations.
Goitom is originally from Ethiopia. He graduated with an Erasmus Mundus Joint Master’s degree in Intelligent Field Robotic Systems from the University of Girona, Spain, and Eötvös Loránd University, Hungary, where he gained research and hands-on experience in simultaneous localization and mapping, sensor fusion, motion planning, and machine learning for autonomous robotic systems. His thesis focused on visual-inertial SLAM, where he worked on accurate localization and 3D mapping of autonomous robots operating under different environmental conditions.
With a strong background in robotics, machine learning, and embedded systems, his research focuses on split computing and early-exit techniques.