Welcome! Iām Jed Guzelkabaagac, an MSc student in Mathematics in Data Science at the Technical University of Munich, where I am also a research scholar with the Konrad Zuse School of Excellence in Reliable AI (relAI).
Iām interested in the mathematics and foundations of machine learning, with an emphasis on generative models, robustness, and graph-based learning.
- š CV
- š Google Scholar
- š Projects
- š» GitHub
- āļø Email
Highlights
š” ICML 2025 Spotlight
Co-author on Privacy Amplification by Structured Subsampling for Deep Differentially Private Time Series Forecasting, accepted as a spotlight paper (top 2.6% of submissions).
Co-author on Privacy Amplification by Structured Subsampling for Deep Differentially Private Time Series Forecasting, accepted as a spotlight paper (top 2.6% of submissions).
š Academic Excellence
M.Sc. GPA 1.2 (top 2% of cohort), Technical University of Munich.
M.Sc. GPA 1.2 (top 2% of cohort), Technical University of Munich.
š¬ Research Experience
Projects spanning robustness, generative models, uncertainty estimation, and scientific ML. Workshop submission in robotics in progress. See portfolio ā
Projects spanning robustness, generative models, uncertainty estimation, and scientific ML. Workshop submission in robotics in progress. See portfolio ā
š International Research
Experience across the UK, Germany, and Canada, including a UARE placement at the University of Alberta.
Experience across the UK, Germany, and Canada, including a UARE placement at the University of Alberta.
š
Scholarships
- Konrad Zuse School of Excellence in Reliable AI (2023ā25)
- University of Alberta Research Experience Stipend (2025)
- Deutschlandstipendium (2023ā24, declined)
- Konrad Zuse School of Excellence in Reliable AI (2023ā25)
- University of Alberta Research Experience Stipend (2025)
- Deutschlandstipendium (2023ā24, declined)