Forschung

• Decision Making under Uncertainty
• Inverse Reinforcement Learning from Human Behavior
• Stochastic Optimization

Lehre

• Übungen zu Grundlagen der Mess- und Regelungstechnik
• Tutorial for Measurement and Control Systems
• Übungen zu Regelungstechnik und Systemdynamik

Zu vergebende Abschlussarbeiten

• Learning Driving Behavior using Inverse Reinforcement Learning from Failures
• Combining Monte Carlo Tree Search with Reinforcement Learning for Decision Making under Uncertainty
• Efficiently solving POMDPs with constant hidden states by solving an equivalent family of MDPs
• Decision Making under Uncertainty with MCTS-based online POMDP planning
• Deep Inverse Reinforcement Learning for Lane Changes on Highways

Veröffentlichungen

Danial Kamran, Tizian Engelgeh, Marvin Busch, Johannes Fischer, Christoph Stiller. Minimizing Safety Interference for Safe and Comfortable Automated Driving with Distributional Reinforcement Learning. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, September 2021. accepted.

Johannes Fischer, Christoph Eyberg, Moritz Werling, Martin Lauer. Sampling-based Inverse Reinforcement Learning Algorithms with Safety Constraints. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, September 2021. accepted.

Johannes Fischer, Ömer Sahin Tas. Information Particle Filter Tree: An Online Algorithm for POMDPs with Belief-Based Rewards on Continuous Domains. In Proceedings of the 37th International Conference on Machine Learning (ICML), Vienna, Austria, July 2020. [ .html ]