Senior PI at Helmholtz AI; Associate Professor of Algorithmic Machine Learning & Explainable AI at TU Munich

Prof. Dr. Stefan Bauer

We work on all phases of experimental design and we want to automate data collection and make data collection processes significantly more efficient!

Group Page 

Academic Career and Research Areas

Stefan Bauer is an Associate Professor of Algorithmic Machine Learning & Explainable AI at TU Munich and Senior PI at Helmholtz AI. He joined Helmholtz Munich on March 1, 2023, returning to his home region of Munich after an international research career. Stefan studied Mathematics at ETH Zurich and Economics and Finance at the University of London before completing his Ph.D. in Computer Science at ETH Zurich in 2018, for which he received the prestigious ETH Medal for outstanding dissertations. After his doctorate, he became a Group Leader at the Max Planck Institute for Intelligent Systems in Tübingen and later an Assistant Professor at KTH Stockholm. He also held visiting researcher positions at MILA (Montreal), GSK, and Microsoft Research. 

His research focuses on developing AI systems that learn causal relationships from high-dimensional data, can explain their decisions, and adapt rapidly to new tasks and environments. These capabilities are key to building robust, trustworthy AI with real-world impact. 

At Helmholtz Munich, Stefan collaborates across disciplines – from single-cell biology to protein design – to apply and advance causal machine learning. He co-coordinates the Helmholtz Foundation Model Initiative, one of Germany’s major national efforts for developing large-scale AI models, and is a CIFAR LMB Fellow (since 2025) and former CIFAR Azrieli Global Scholar. His publications have received multiple awards, including the Best Paper Award at ICML 2019.

Fields of Work and Expertise

Representation Learning

Foundation Models

Reinforcement Learning

Experimental Design

Algorithmic Machine Learning

Causal Inference

In the recent Federal Agency for Breakthrough Innovation (Bundesagentur für Sprunginnovation, SPRIND) competition, 2 out of 8 competitively selected (of the overall 150) and generously funded startup teams are affiliated with the Bauer lab.

Professional Background

2018

Ph.D. in Computer Science, ETH Zurich (awarded ETH Medal for outstanding dissertation)

2018 – 2022

Group Leader, Max Planck Institute for Intelligent Systems, Tübingen; Visiting Researcher at MILA, GSK, and Microsoft Research

2022 – 2023

Assistant Professor, KTH Royal Institute of Technology, Stockholm

2023 – Present

Associate Professor, TU Munich & Senior PI at Helmholtz AI / Helmholtz Munich (joined March 1)

2025 – Present

CIFAR LMB Fellow; Co-coordinator of the Helmholtz Foundation Model Initiative (HFMI)

Honors and Awards

  • ETH Medal for Outstanding Dissertation, ETH Zurich (2018);
  • CIFAR Azrieli Global Scholar (previously); CIFAR LMB Fellow (since 2025);
  • Best Paper at main ML conferences and workshops (ICML, CVPR);

 

 

Most Recent Publications

Nat. Comput. Sci. 5, 801–812 (2025)

Wei, Y. ; Peng, B. ; Xie, R. ; Chen, Y. ; Qin, Y. ; Wen, P.Y. ; Bauer, S. ; Tung, P.Y. ; Raabe, D.

Deep active optimization for complex systems.
Trans. Machine Learn. Res. 2025, accepted (2025)

Angelis, E. ; Quinzan, F. ; Soleymani, A. ; Jail, P.J. ; Bauer, S.

Double machine learning based structure identification from temporal data.
Diabetologia 68, 2277-2289 (2025)

Parajuli, A. ; Bendes, A. ; Byvald, F. ; Stone, V.M. ; Ringqvist, E.E. ; Butrym, M. ; Angelis, E. ; Kipper, S. ; Bauer, S. ; Roxhed, N. ; Schwenk, J.M. ; Flodstrom-Tullberg, M.

Frequent longitudinal blood microsampling and proteome monitoring identify disease markers and enable timely intervention in a mouse model of type 1 diabetes.

Xian, R.P. ; Baker, N.R. ; David, T. ; Cui, Q.H. ; Holmgren, A.J. ; Bauer, S. ; Sushil, M. ; Abbasi-Asl, R.

Robustness tests for biomedical foundation models should tailor to specifications.
2025 in
In: (13th International Conference on Learning Representations Iclr 2025, 24 - 28 April 2025, Singapur). 2025. 28433-28468

Mamaghan, A.M.K. ; Papa, S. ; Johansson, K.H. ; Bauer, S. ; Dittadi, A.

Exploring the effectiveness of object-centric representations in visual question answering: Comparative insights with foundation models.
Futuristic medical concept with red human lungs. Abstract geometric design with plexus effect on dark background. Healthcare and pulmonology banner with copy space.

Helmholtz AI

Democratising AI for a data-driven future


We democratize AI and advance its capabilities for tackling frontier scientific questions while harnessing its power for solving the grand challenges facing society. We are an application-driven artificial intelligence platform accelerating science across the Helmholtz Association. We enable the development and implementation of AI solutions while promoting collaboration and ensuring accessibility to resources and expertise.

To Website