Michael Kamp

Michael Kamp is an Associate Professor for Machine Learning and Artificial Intelligence at TU Dortmund University and a faculty member of the Lamarr Institute for Machine Learning and Artificial Intelligence. He maintains affiliations with the Institute for Artificial Intelligence in Medicine (IKIM) at University Medicine Essen and collaborates with institutions in the UA Ruhr consortium.

• Formerly led Trustworthy Machine Learning group at IKIM
• Postdoctoral researcher at CISPA Helmholtz Center (2021) and Monash University (2019-2021)
• Doctorate from University of Bonn

His research focuses on trustworthy machine learning through three pillars: deep learning theory (loss surfaces, flatness, generalization), causal representation learning for explainability, and federated learning for privacy-preserving decentralized systems. Applications span healthcare, autonomous driving, and cybersecurity.

Recent publications at AAAI/ICLR/NeurIPS explore: federated optimization algorithms for non-IID data, privacy-preserving causal discovery, and flatness-aware regularization techniques. His work combines theoretical rigor with real-world deployment challenges.

• NeurIPS 2021 Outstanding Reviewer Award
• ICLR 2021 Reviewer Award
• Best Paper at PDFL'20 Workshop
• NeurIPS 2019 Best Reviewer Award
• ICML 2019 Reviewer Award

He contributes to the editorial board of Springer's Machine Learning journal and serves in the ELLIS society. His group's tools address early disease detection, 3D medical shape analysis, and domain transfer challenges while maintaining ethical constraints.