activity
20232026
collaborators

7 papers

math.AG2026

Fast Isotopy Computation for T-Curves

Zoe Geiselmann, Michael Joswig, Lars Kastner +5

A T-curve of degree is given by a regular unimodular triangulation of together with a sign distribution on its lattice points. By Viro's Patchworking Theorem, thi…

math.AG2026

Limits of combinatorial patchworking

Zoe Geiselmann, Michael Joswig, Lars Kastner +5

It is shown that there are real plane algebraic curves of degree eight that cannot be realized as T-curves, i.e., via combinatorial patchworking. In fact, this holds for several re…

cs.LG2025

Neural Sum-of-Squares: Certifying the Nonnegativity of Polynomials with Transformers

Nico Pelleriti, Christoph Spiegel, Shiwei Liu +3

Certifying nonnegativity of polynomials is a well-known NP-hard problem with direct applications spanning non-convex optimization, control, robotics, and beyond. A sufficient condi…

cs.LG2025

Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?

Konrad Mundinger, Max Zimmer, Aldo Kiem +2

We demonstrate how neural networks can drive mathematical discovery through a case study of the Hadwiger-Nelson problem, a long-standing open problem at the intersection of discret…

math.CO2024

Extending the Continuum of Six-Colorings

Konrad Mundinger, Sebastian Pokutta, Christoph Spiegel +1

We present two novel six-colorings of the Euclidean plane that avoid monochromatic pairs of points at unit distance in five colors and monochromatic pairs at another specified dist…

math.CO2023

The Four-Color Ramsey Multiplicity of Triangles

Aldo Kiem, Sebastian Pokutta, Christoph Spiegel

We study a generalization of a famous result of Goodman and establish that asymptotically at least a fraction of all triangles needs to be monochromatic in any four-colorin…

cs.LG2023

PERP: Rethinking the Prune-Retrain Paradigm in the Era of LLMs

Max Zimmer, Megi Andoni, Christoph Spiegel +1

Neural Networks can be effectively compressed through pruning, significantly reducing storage and compute demands while maintaining predictive performance. Simple yet effective met…