M.Sc. Pascal Zimmer

About Me

Since January 2022, I am a PhD student at the Chair for Information Security. Before that, I worked as a research assistant at the Chair for Embedded Security at the Ruhr University Bochum and as a visiting researcher at the Embedded Security group of the Max Planck Institute for Security and Privacy. I obtained both my Bachelor’s and Master’s degree at the Ruhr University Bochum in cybersecurity with a focus on wireless physical layer security. 

Research

My research focuses on the impact and robustness of machine learning systems. This includes but is not limited to MLaaS deployments, distributed and decentralized learning systems, e.g., federated learning, and real-world deployments, such as autonomous vehicles or facial recognition systems.

Publications

• Pascal Zimmer, Sebastien Andreina, Giorgia Marson, Ghassan Karame,
  Closing the Gap: Achieving Better Accuracy-Robustness Tradeoffs Against Query-Based Attacks,
  Proceedings of the AAAI Conference on Artificial Intelligence, 38(19), 21859-21868, 2024. [ PDF | BibTex ]

theses supervision

I had the pleasure of co-supervising theses of many talented students in the field of machine learning security. If you are interested in writing a thesis with us, have a look at this page.

• Pablo Schmücker, „Evaluating defense mechanisms against universal adversarial perturbations for convolutional neural networks“, B.Sc.
• Luis Griepenstroh, „Mitigation Strategies for Real-World Adversarial Patches“, B.Sc.
• Max Randhahn, „Security Analysis of Adversarial Attacks in Federated Learning“, B.Sc.
• Jan Richter, „Security Analysis of Transfer-based Adversarial Attacks in the Graybox Model“, M.Sc.
• Alexander Jan Zielinski, „Imperceptible Adversarial Examples for Autonomous Vehicles“, M.Sc.
• Fabian Rüsen, „Security Analysis of Adversarial Examples in the Graybox Model“, M.Sc.
• Pascal Bongartz, „Cross-Dimensional Security Assessment of an AutoML Training Pipeline“, M.Sc.