Research Interests

My research focuses on semantic and explainable AI (XAI), as well as the robustness of AI systems. I am particularly interested in methods such as Layer-wise Relevance Propagation (LRP) and other explainability techniques, Transformer-based models for a wide range of applications, and CLIP and related zero-shot learning approaches.

My main application domains include Medical AI and other safety-critical areas, where transparency and reliability of AI systems are essential.

Topics for Theses

If you are interested in writing a thesis related to my research areas, feel free to contact me via email.

Most thesis topics are centered around AI and machine learning, so please include a short note about your experience with AI development, this helps me suggest a topic that best fits your background.

If you are currently doing a Bachelor’s degree, no prior AI knowledge is mandatory. However, for a Master’s thesis, I recommend having at least a basic understanding of AI principles.

So even if you’re new to AI, don’t worry, we’ll find a topic that works for you! 😊

Supervised Theses

---

Publications (2)

2025 (2)

• Singerhoff, Malte and Weis, Torben
  Imputation Matters: Evaluating the Impact of Missing Data Strategies on Interpretability in Clinical Time Series Models
  In: UbiComp/ISWC 2025 Adjunct Proceedings.
  DOI: 10.1145/3714394.3756301
• Singerhoff, Malte
  XTRTimeS — eXplainable Transformer for Robust Time Series Forecasting
  In: Proceedings of the 23rd IEEE International Conference on Pervasive Computing and Communications (PerCom 2025)
  DOI: 10.1109/PerComWorkshops65533.2025.00144

---

Teaching

WiSe 25/26

• CPS-Labor Praktikum
• Distributed Systems

SoSe 2025

• CPS-Labor Praktikum
• Internet-Technologie & Web Engineering

WiSe 24/25

• Distributed Systems

SoSe 2024

• Internet-Technologie & Web Engineering