Marius Köppel

Particle Physicist | Fairness-Preserving Machine Learning

marius_art.png

I am a PostDoc in Partical Physics at the ETH Zurich, working on Experimantal Partical Physics, Real Time Data Acquisition and Machine Learning. At ETH I am responsible for the Scintillating Fibre detector of the Mu3e and working on ML based anomaly detection for the CMS experiment. My research has the focus on developing solutions that can handle the vast amounts of data generated by scientific experiments today.

I did my PhD in Physics at the Johannes Gutenberg Universität Mainz, particularly developing the data acquisition system for the Mu3e experiment.

The Mu3e experiment is designed to search for the rare decay of the muon (\(\mu^+ \rightarrow e^+ + e^+ + e^-\)), leading to a significant step towards understanding the fundamental laws the universe.

Furthermore, I am actively interested in enhancing detector technologies in general. Therefore, I am working on doing Muon Spin Spectroscopy using Si-Pixel detectors (MuSiP) deployed at the Paul Scherrer Institute.

My work also extends beyond the boundaries of physics, venturing into the field of algorithmic fairness. This domain is where ethics meets technology, challenging us to design and implement algorithms that are not only efficient but also equitable. I am an associate member of the TOPML project, which studies the interaction of non-functional properties in machine learning.

selected publications

  1. binarymi.png
    Invariant Representations with Stochastically Quantized Neural Networks
    Mattia Cerrato , Marius Köppel , Roberto Esposito , and 1 more author
    Proceedings of the AAAI Conference on Artificial Intelligence, Jun 2023
  2. dataflow.png
    Data Flow in the Mu3e DAQ
    Marius Köppel
    IEEE Transactions on Nuclear Science, Jun 2023
  3. 10y.png
    10 Years of Fair Representations: Challenges and Opportunities
    Mattia Cerrato , Marius Köppel , Philipp Wolf , and 1 more author
    Jun 2024