Research Interest
Having a strong background in computer vision, my research focuses on decision-making under the uncertainty inherent in visual perception. I am particularly interested in developing safe reinforcement learning agents that utilize object recognition to perceive their dynamic environment. My work aims to bridge the gap between perception and action by exploring how intelligent systems can make robust decisions despite uncertain perceptions, ensuring safety and reliability in real-world scenarios.
Supervised Theses
- Vath, Phillip (2025). Uncertainty-Aware Perception for Reinforcement Learning Agents
- Höner, Laurenz (2025). On Feature Space Reduction and State Space Discretization: A Pipeline for Vision-Based Reinforcement Learning
- Kamphausen, Christopher (2025). Adversarial Attacks on the Visual Perception of Reinforcement Learning Agents
- Wagner, Tim (2025). A Pipeline for Uncertainty Estimation in the Visual Perception of Reinforcement Learning Agents
- Andreou, Nikos (2025). Leveraging Multi-Stage Reasoning for Trend Detection and Cluster Identification
- Kachanov, Andreas (2025). Quantifying Uncertainties in Depth Estimation for Autonomous Driving in CARLA.
Publications
TBD