Research

I joined the Chair of Artificial Intelligence and Machine Learning in July 2025. My research focuses on machine learning for physics, particularly neural operators, dynamical systems, and combining numerical simulations with data-driven methods.
I’m especially interested in neural operators as a means to efficiently model complex physical systems by learning mappings between function spaces. My goal is to embed physical knowledge—such as symmetries or conservation laws—into neural architectures to improve generalization and efficiency. This approach can lead to more accurate weather forecasts, faster mechanical simulations, and improved protein folding predictions.
More broadly, I’m also interested in geometric deep learning, graph-based methods, and time-series analysis.

Background

I studied Mathematics for my Bachelor's and Master's degrees. I'm funded by the Munich Center for Machine Learning (MCML) and am a member of the ELLIS PhD program.