Turan Orujlu

PhD Student



I’m interested in intuitive physics for artificial intelligence systems. I hope to contribute to the field throughout my PhD studies, with a specific focus on compositional scene understanding of embodied agents via object-centric latent representations. Understanding a scene goes beyond simply labelling objects in a scene and their locations, it should involve creating a model of the environment that lets the agent act in accordingly. I believe modelling object interactions from the perspective of the agent is the necessary first step to achieving this goal, and would hope to work on this problem as part of my PhD project.

Before embarking on this PhD journey, I’ve received BA in Mathematical Economics from the University of Pennsylvania, BSc in Cognitive Science from the University of Osnabrück, and MSc in Neural Information Processing from the University of Tübingen.