Machine Learning Group

The Machine Learning Group at TU Berlin is chaired by Prof. Dr. Klaus-Robert Müller and focuses on methodological and theoretical improvements as well as applications in machine learning. As for methodology, some of our research areas are explainable AI, probabilistic ML, and multimodal learning. As for applications, exemplary domains are quantum chemistry, digital pathology, and biomedical sensing.

Recent Publications

Explaining Predictive Uncertainty by Exposing Second-Order Effects.
Preemptively Pruning Clever-Hans Strategies in Deep Neural Networks.

News

Feb 2024: The AQTIVATE Workshop kicks off on Monday, Feb 12th 2024, at Technische Universität Berlin, offering an immersive dive into Machine Learning with a focus on applications in science. Organized by BIFOLD and the Machine Learning Group, the workshop covers everything from basic principles to specialized uses in physics and chemistry. Further details and updates are available here.