About Me
I am a post-doctoral researcher at the Max-Planck Institute for Mathematics in the Sciences in Leipzig, supervised by Guido Montufar and Bernd Sturmfels. I completed my PhD in Mathematics at TU Berlin under the supervision of Martin Skutella. My research focuses on neural networks theory using polyhedral geometry, combinatorics, computational complexity and topology. I previously earned a B.Sc. in Mathematics and Computer Science and an M.Sc. in Mathematics from the University of Greifswald.
News
- Two new preprints regarding identifiability and fibers of ReLU networks are now available: Most ReLU Networks Admit Identifiable Parameters and The Symmetries of Three-Layer ReLU Networks.
Publications
Parameterized Hardness of Zonotope Containment and Neural Network Verification
International Conference on Learning Representations (ICLR) 2026
Depth-Bounds for Neural Networks via the Braid Arrangement
(Oral) Conference on Neural Information Processing Systems (NeurIPS) 2025
Complexity of Deciding Injectivity and Verification of ReLU Neural Networks
Conference on Learning Theory (COLT) 2025
Decomposition Polyhedra of Piecewise Linear Functions
(Spotlight) International Conference on Learning Representations (ICLR) 2025
Conference Version | Preprint | Video | Poster | Slides
Topological Expressivity of ReLU Neural Networks
Conference on Learning Theory (COLT) 2024
Thesis
Expressivity and Complexity of ReLU Neural Networks