Bilkent University
Department of Computer Engineering


Physical reasoning: How to go from seeing to acting


Ozgur S. Oguz
Learning and Intelligent Systems Lab, TU Berlin

Autonomous robots are envisioned to be ubiquitous in our daily lives. Such robots are expected to make sequential decisions, plan their motions, and control their movements to realize their tasks. This remarkable skill set requires a new research direction where perception, discrete decision-making, motion planning, control, and learning methods are considered jointly to provide autonomy to the agent while physically interacting with the world. In this talk, I will present our initial steps toward tackling this challenge. I will cover our research on (i) explainable and effective representations directly from visual perception data in detail. To provide a broader perspective, I will also highlight the keypoints of our work on (ii) task decompositions and robust motion planning algorithms for long-horizon tasks, and (iii) (safe) learning for control of autonomous robots in the real-world.

Bio: Ozgur S. Oguz received the B.Sc. and M.Sc. degrees in computer science from Koc University, Istanbul, Turkey, and the Ph.D. degree in robotics (summa cum laude) from the Department of Electrical and Computer Engineering, Technical University of Munich, Germany, in 2018. He was a Postoctoral Researcher at TU Munich, and Max Planck Institute for Intelligent Systems. He is currently a Postdoctoral Researcher with the Learning and Intelligent Systems Lab, TU Berlin, and affiliated with the Cluster of Excellence Integrative Computational Design and Construction for Architecture (IntCDC) at University of Stuttgart. His research interests include developing autonomous systems that are able to reason about their states of knowledge, make sequential decisions to realize a goal, and simultaneously learn to improve their causal physical reasoning and manipulation skills.


DATE: 01 November 2021, Monday @ 11:30