David Schnurr

I am currently pursuing a Master of Science in Computer Science at ETH Zurich, specializing in Machine Intelligence with a Minor in Data Management. My main research interests lie in tabular machine learning, time series forecasting and addressing (temporal) distribution shifts.

Prior to joining ETH, I completed my Bachelor of Science in Computer Science at the University of Freiburg, where I graduated with a grade average of 1.1 (top 2%) and a focus on Machine Learning complemented with Algorithm Development and Analysis.

I have gained practical experience through a 7-month internship at Rheinmetall Air Defence AG, where I contributed to improving a real-time computer vision pipeline. This role allowed me to apply my deep learning and software engineering skills to optimize latency and accuracy while improving the system's robustness. I also led efforts to integrate our module with other systems, collaborating closely with multiple departments.

Additionally, I worked as a research assistant at the Machine Learning Lab of the University of Freiburg. There, I enhanced the TabPFN model by implementing techniques such as comprehensive missing value generation and model fine-tuning, resulting in improved benchmark performance.

Email  /  LinkedIn  /  Google Scholar  /  Github

profile photo

Publications


Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Data
Kai Helli*, David Schnurr*, Noah Hollmann, Samuel Müller, Frank Hutter; * Equal contribution
NeurIPS 2024 (Main Track) ; TRL @ NeurIPS 2024 ; AutoML Conference 2024 (Workshop Track)