Stefan Heidekrüger
	Department of Informatics (I18) Technical University of Munich
	E-Mail:	stefan heidekrueger   tum de           .                    @       .
	Office:	Room 01.10.056 Boltzmannstr. 3 85748 München, Germany
	Phone:	+49 (0) 89 289 - 17530
	Hours:	by arrangement

I'm a PhD student at the DSS chair supervised by Prof. Bichler. My research focusses on computation of equilibria in markets and auctions via multi-agent reinforcement learning methods.

We currently have several possible topics for student projects (MSc or BSc theses, IDP, Guided Research) in this research area. If you're looking for a project and interested in (Multi-Agent) Reinforcement Learning, Neural Networks, Nonlinear Optimization, Market Design and/or Algorithmic Game Theory, feel free to get in touch.

Publications

S. Heidekrüger, P. Sutterer, and M. Bichler. Computing approximate bayes-nash equilibria through neural self-play. In Workshop on Information Technology and Systems (WITS19), Munich, Germany, 2019.

Conference Talks

• Computing approximate Bayes-Nash equilibria through Neural Self-Play (Workshop on Information Technologies and Systems, Munich, Germany, 12/2019)
• Computing Bayes-Nash Equilibria in Sealed Bid Auctions through Self Play (International Meeting on Operations Research, Dresden, Germany, 09/2019)

Teaching

Courses

• WS 18/19, 19/20     Business Analytics, Teaching Assistant
• SS 19                      Seminar on Data Mining, TA          

Completed Student Projects

• Kevin D. Falkenstein    Learning Equilibrium Strategies in Auctions via Deep Neural Networks, MSc Thesis (2018)
• Sebastian Rief              Detection of anomalies in large-scale accounting data using unsupervised machine learning, MSc Thesis (2019)
• Florian Ziesche             Human Interpretable Machine Learning: A Machine Learning Approach for Risk Scoring, MSc Thesis (2019)