Maurice Kraus

Machine Learning Group, Computer Science Department, TU Darmstadt.
Hochschulstrasse 1, Building S1|03, Room 077, 64289 Darmstadt, Germany.
maurice (dot) kraus (at) cs (dot) tu-darmstadt (dot) de

My research explores how deep learning models can learn from complex, evolving data over time—often involving multiple sources of information.

I am broadly interested in methods that make sequence and forecasting models more efficient and, therefore, more useful in practice. This includes questions about how models capture temporal structure, how they generalize under changing conditions, and how we can better understand and evaluate their behavior.

The overarching goal is to develop learning systems for time-dependent data that can be applied with greater confidence.

I regularly supervise student projects and theses in time-series forecasting, efficient sequence modeling, explainability/robustness, and benchmarking. If you are interested, please email me (i) a short description of your interests, (ii) your CV and transcript, and (iii) your preferred start date. I am also happy to discuss your own topic proposals.