Michael Kirchhof
Collaborating PhD Researcher
Github Linkedin michael.kirchhof(at)uni-tuebingen.de Google Scholar


Michael Kirchhof is a Ph.D. candidate in the International Max-Planck Research School for Intelligent Systems (IMPRS-IS), co-supervised by Prof. Enkelejda Kasneci and Prof. Zeynep Akata at the University of Tübingen. He received his B. Sc. (2018) and M. Sc. (2021) in Statistics with distinction at TU Dortmund University where he focussed on probabilistic modeling and machine learning. In 2019, he was a research intern at BMW Group, Munich. His research goal is to introduce explanations to safety-critical classification tasks. To this end, he is interested in the structure of learned representations and the dualities between probability theory and machine learning.


  • "A Non-isotropic Probabilistic Take on Proxy-based Deep Metric Learning", Michael Kirchhof *, Karsten Roth *, Zeynep Akata, Enkelejda Kasneci European Conference on Computer Vision, ECCV 2022