Hi, I am Michael

NEWS My PhD is over and I started looking for jobs. Feel free to contact me.

NEWS Paper accepted at NeurIPS 2022. Read it here!

Currently, the main topic of my PhD is fairness in machine learning which considers the impact of automated decision-making on human lifes. In order to make sure good decisions are made, everyone applies some notion of fairness to a large-scale decision-making system, in particular, when these decisions change people’s lives. I study decision-making systems, which aim to be fair to groups of individuals, and I focus on investigating their fairness properties.

I have also studied Comparison-based learning, where we are not given representations of data points, but we know for three items A, B, C if item A is closer to item B or closer to item C.

In October 2017, I started my PhD in the Theory of Machine Learning group (TML) supervised by Prof.Dr. Ulrike von Luxburg at the University of Tübingen. I am a scholar in the International Max Planck Research School for Intelligent Systems (IMPRS-IS), a graduate school by both the universities and Max-Planck-Institutes in Tuebingen and Stuttgart.

Interests

  • Comparison-Based Machine Learning
  • Fairness in Machine Learning

Education

  • Ph.D. in Machine Learning, starting 2017

    University of Tübingen

  • MSc in Mathematics and Economics, 2017

    University of Cologne

  • BSc in Mathematics and Economics, 2014

    University of Cologne

Publications

Are Two Heads the Same as One? Identifying Disparate Treatment in Fair Neural Networks

We show that deep neural networks that satisfy demographic parity do so through a form of race or gender awareness, and that the more …
PDF

Leveling Down in Computer Vision: Pareto Inefficiencies in Fair Deep Classifiers

Algorithmic fairness is frequently motivated in terms of a trade-off in which overall performance is decreased so as to improve …
PDF

Insights into Ordinal Embedding Algorithms: A Systematic Evaluation

The objective of ordinal embedding is to find a Euclidean representation of a set of abstract items, using only answers to triplet …
PDF

Too Relaxed to Be Fair

We address the problem of classification under fairness constraints. Given a notion of fairness, the goal is to learn a classifier that …
PDF Code Video/Slides

Uncertainty Estimates for Ordinal Embeddings

To investigate objects without a describable notion of distance, one can gather ordinal information by asking triplet comparisons of …
PDF

Important to Me

Student Exchange. In 2008/2009 I spent a whole year in Maryland in the U.S. as an exchange student because I was granted the PPP scholarship from the Bundestag. I encourage everyone to apply here!

I was very lucky to be supported by Partnership International e.V. during my exchange year. Ever since I came back, I was part of PI and we have done lots of voluntary work together. If you are interested in hosting an exhange student yourself, or if your children are interested in going on an exchange, check out our website.

Education. Every summer students interested in growing and learning come together for three weeks and they code, they create, sing, dance, … simply do whatever challenges them. Two times, I offered a course on ML and Renewable Energy, and once, I organized the overall academy. I can only recommend to check it out, either as a high-school student or as a university student or PhD for offering courses: https://www.schuelerakademien.de.

Contact