Dr. Michael Oesterle

Postdoc

Universität Mannheim
Institut für Enterprise Systems
L 15, 1–6
L15, 1–6 – Raum 417
68161 Mannheim

E-Mail: michael.oesterleuni-mannheim.de

Forschungsinteressen

Mechanistische Interpretierbarkeit von Machine Learning-Modellen, insbesondere Transformer-Architekturen
Steuerung und Koordination von Multiagentensystemen mithilfe von Spieltheorie, Mechanism Design und Reinforcement Learning

Lebenslauf

Seit 2024	Wissenschaftlicher Mitarbeiter und Postdoc am Institut für Enterprise Systems
2022	Fulbright-Forschungsaufenthalt an der Texas A&M University (TAMU)
2018–2024	Wissenschaftlicher Mitarbeiter und Doktorand am Institut für Enterprise Systems
2014–2017	Consultant bei der TWS Partners AG
2012–2014	Master of Science Mathematics in Operations Research an der Technischen Universität München (TUM)
2009–2012	Bachelor of Science Mathematik an der Technischen Universität München (TUM)

Publikationen

Oesterle, M., Grams, T. und Bartelt, C. (2024). DRAMA at the PettingZoo: Dynamically restricted action spaces for multi-agent reinforcement learning frameworks. In , Proceedings of the 57th Annual Hawaii International Conference on System Sciences, HICSS 2024, Hilton Hawaiian Village Waikiki Beach Resort, Hawaii, USA, January 3–6, 2024 (S. 7810-7819). , Department of IT-Management, Shidler College of Business, University of Hawaii: Honolulu, HI.
Oesterle, M., Grams, T., Bartelt, C. und Stuckenschmidt, H. (2024). RAISE the bar: Restriction of action spaces for improved social welfare and equity in traffic management. In , Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems (S. 1492-1500). , International Foundation for Autonomous Agents and Multiagent Systems: Richland, SC.
Rana, A., Oesterle, M. und Brinkmann, J. (2024). GOV-REK: Governed Reward Engineering Kernels for designing robust multi-agent reinforcement learning systems. In , AAMAS '24: proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems (S. 2429-2431). , International Foundation for Autonomous Agents and Multiagent Systems: Richland, SC.
Oesterle, M. und Sharon, G. (2023). Socially optimal non-discriminatory restrictions for continuous-action games. In , Proceedings of the 37th AAAI Conference of Artificial Intelligence. Vol. 10 (S. 11638-11646). , AAAI Press: Washington, DC.
Oesterle, M. und Sharon, G. (2023). Socially optimal non-discriminatory restrictions for continuous-action games. In , KI 2023: Advances in Artificial Intelligence : 46th German Conference on AI, Berlin, Germany, September 26–29, 2023, Proceedings (S. 252–256). Lecture Notes in Computer Science, Springer: Berlin [u.a.].
Oesterle, M., Bartelt, C., Lüdtke, S. und Stuckenschmidt, H. (2022). Self-learning governance of black-box multi-agent systems. In , Coordination, Organizations, Institutions, Norms, and Ethics for Governance of Multi-Agent Systems XV : International Workshop, COINE 2022, virtual event, May 9, 2022, revised selected papers (S. 73–91). Lecture Notes in Computer Science, Springer: Berlin [u.a.].
Pernpeintner, M. (2021). Self-learning governance of competitive multi-agent systems. In , Organic Computing : Doctoral Dissertation Colloquium 2020 (S. 47–63). Intelligent Embedded Systems, Kassel University Press: Kassel.
Pernpeintner, M. (2021). Toward a self-learning governance loop for competitive multi-attribute MAS. In , AAMAS '21: Proceedings of the 20th International Conference on Autonomous Agents and MultiAgent Systems : Richland, SC, Virtual Event United Kingdom, May, 2021 (S. 1619-1621). , International Foundation for Autonomous Agents and Multiagent Systems: Richland, SC.
Pernpeintner, M., Bartelt, C. und Stuckenschmidt, H. (2021). Governing black-box agents in competitive multi-agent systems. In , Multi-Agent Systems : 18th European Conference, EUMAS 2021, virtual event, June 28–29, 2021, revised selected papers (S. 19–36). Lecture Notes in Computer Science, Springer: Berlin [u.a.].
Pernpeintner, M. (2020). Achieving emergent governance in competitive multi-agent systems. In , AAMAS '20 : Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems, Auckland, Nea Zealand, May 2020 (S. 2204-2206). , ACM Digital Library: New York, NY.
Pernpeintner, M. (2019). Collaboration as an emergent property of self-organizing software systems. In , 2019 IEEE 4th International Workshops on Foundations and Applications of Self* Systems : FAS*W 2019 : proceedings (S. 231–233). , IEEE Computer Society Press: Piscataway, NJ.
De Loera, J. A., Margulies, S., Pernpeintner, M., Riedl, E., Rolnick, D., Spencer, G., Stasi, D. und Swenson, J. (2015). Graph coloring ideals : Nullstellensatz certificates, Gröbner bases for chordal graphs, and hardness of Gröbner bases. In , ISSAC '15 : proceedings of the 2015 ACM International Symposium on Symbolic and Algebraic Computation : July 6–9, 2015, Bath, United Kingdom (S. 133–140). , ACM: New York, NY.

Betreute Abschlussarbeiten

Governance for Sparse Multi-Agent Systems (Masterarbeit, 2023)
Deep Reinforcement Learning in Obstacle Avoidance to Investigate Interval Restrictions on Action Spaces (Masterarbeit, 2022/23)
Application and Evaluation of Multi-Agent Reinforcement Learning Approaches for the Coordinated Routing of Drones (Masterarbeit, 2021/22)
Data Modeling And Processing for AI-based Governance of Competitive Multi-Agent Systems (Masterarbeit, 2020/21)
Building an AI Agent for Strategic Interaction (Bachelorarbeit, 2019)
Coverage-based Testing Improvement in the Domain of automated Driving – Utilizing unsupervised Selection of Traces in Field Data for Test Scenario Reconstruction (Masterarbeit, 2018/19)