I am a postdoctoral researcher at the Tübingen AI center at the University of Tübingen, Germany working with Jakob Macke.
My research is fairly diverse. Mainly, I try to understand deep learning through theory and empirical observations. I am interested in
- low-dimensional data representation learning,
- modelling of temporal data,
- the role of overparameterization,
- unsupervised representation learning.
On the applied side, my research uses deep models for the evaluation of electrocardiogram (ECG) recordings to (1) support physicians with diagnosis and (2) explore the information within those recordings.
Background
- Currently: PostDoc at Tübingen University, Germany.
- 2024: Ph.D. in deep learning at Uppsala University, Sweden.
- 2019: M.Sc. in systems and control from TU Delft, the Netherlands.
- 2015: B.Sc. in aerospace engineering from DHBW, Germany.
- 1994: born.
Latest research results and news
September 15, 2024, Postdoc: I started my postdoc at the Tübingen AI center at the University of Tübingen, Germany.
June 14, 2024, Ph.D.: I successfully defended my Ph.D. thesis “On Deep Learning for Low-Dimensional Representations”.
pdf DiVA slides
May 2, 2024, Accepted Paper: Our high-dimensional analysis paper got accepted to ICML. We study the combination of Principal Component Analysis (PCA) with linear regression on data from a spiked covariance model to provide precise asymptotic guarantees for the risk.
OpenReview ICML code poster
April 26, 2024, Accepted Paper: Our work on integrating Recursive Feature Machines (RFMs) into Gaussian Processes was accepted at UAI. We show that the resulting GP-RFM is a strong alternative to existing SOTA methods for tabular regression problems.
OpenReview code poster
All news can be found here