PhD position in Probabilistic and Interactive Machine Learning at University of Vienna

The position is a fully funded 4-year PhD position at the University of Vienna, Faculty of Computer Science, in the Research Group Data Mining and Machine Learning.

The anticipated focus of this PhD position is on reinforcement learning (RL) and its challenges when applied to complex, open-ended, or poorly defined environments. Two of these challenges are sample inefficiency, often caused by the lack of intelligent, structured exploration, and the "reward engineering" problem, where designing an explicit scalar reward function that captures desired behavior is incredibly difficult or impossible. Furthermore, as AI systems are deployed in more complex environments, the challenges of AI alignment (ensuring systems behave according to human preferences) and constrained learning (adhering to strict safety, legal, or physical boundaries) become important. This position is dedicated to addressing these core challenges by advancing the frontiers of (Inverse) Reinforcement Learning (IRL), exploration, and safe/aligned AI.