Our People

About Me

Model-based reinforcement learning is a data-efficient and fast way for autonomous systems to learn new tasks from scratch. In my work, I explore new learning methods for models and policies to bridge the gap to challenging real-world problems, which potentially exhibit noise, delays, and partially observable states. I'm fascinated by autonomous systems, which learn on their own how to act and interact in and with our complex, daily surroundings.

Research Interest: POMDPs, Dynamics Models, Model-Free vs. Model-Based Reinforcement Learning

My Research Fields

• Reinforcement Learning
• Model-Based Policy Search
• Learning Control
• Model Learning