Ozgur S. Oguz

Ozgur S. Oguz is an Assistant Professor in the Department of Computer Engineering at Bilkent University, leading the Learning for Intelligent Robotic Agents (LiRA) Lab. His research focuses on developing algorithms for autonomous agents, emphasizing learning, reasoning, and planning capabilities in robotics and decision-making domains. Prior to joining Bilkent, he was a postdoctoral researcher at the Max Planck Institute for Intelligent Systems and Stuttgart University, and earned his PhD from TU Munich in Human-Robot Interaction (HRI). He holds degrees from UBC and Koç University.

His research interests span AI, robotics, reinforcement learning, and robotic manipulation planning. Notable contributions include work on curriculum-based reinforcement learning, safe learning frameworks for dynamical systems, and hybrid planning algorithms for sequential manipulation tasks. He actively teaches courses such as Learning for Robotics (CS 449/549) and Artificial Intelligence (CS 461).

Recent publications highlight advancements in diffusion-based curriculum learning, corrected experience replay methods, and neural field representations for articulated object manipulation. His work frequently appears in top venues like NeurIPS, ICRA, and IROS. He mentors graduate and undergraduate students in AI and robotics, with current opportunities for postdoctoral researchers.

Key collaborations include projects with the Cluster of Excellence IntCDC and the LiRA Lab’s focus on interdisciplinary robotics research.