Johannes Carmesin
Professor of Discrete Structures · TU Freiberg
The Discrete Structures group at TU Freiberg works on a broad range of topics in graph theory and related areas, including topology and theoretical computer science.

Research team
- Dr Matthew Kroeker
- Dr Utku Okur
- Will J. Turner
- Vaibhav Bajpai
- Chau Do (jointly supervised with C. Brause and I. Schiermeyer)
Research interests

Combinatorics in 3 dimensions
A key goal here is to extend the fundamental methods from Structural Graph Theory to study 2-dimensional simplicial complexes. The starting point for this project is my 3-dimensional analogue of Kuratowski's theorem.

AI and mathematics
We study what AI systems can and cannot do in mathematics. We develop benchmarks in Lean, where proposed proofs can be checked automatically, to assess capabilities ranging from producing formal proofs of established results to tackling open research problems. Our work includes LPBench, a formal-to-formal benchmark for evaluating AI systems on research-level mathematics, and blog posts on the capabilities and limitations of current AI systems, written for a broad mathematical audience.

Graph minors and connectivity
A minor of a graph is obtained by deleting and contracting edges. The connection of Graph Minor Theory with topology is already apparent from this definition as deletions and contractions are "dual operations" for plane graphs, and even more so in the Graph Minor Structure Theorem of Robertson and Seymour . Here I am particularly interested in studying connectivity and tree-decompositions, and using its methods in other areas. New approaches include "local separators" of graphs as well as "angry theorems" for 3-connected graphs.

Matroids
A fundamental theorem in Matroid Theory is Whitney's characterisation of graph planarity in terms of matroids. I extended this theorem to 3-dimensional space. Another direction of research is to extend Whitney's theorem to surfaces, which led to a characterisation of graphs admitting locally planar embeddings.

Infinite graphs
An important tool for studying infinite graphs is the notion of ends, which can be seen as boundary points at infinity of the graph. In my PhD, I proved Halin's end-faithful spanning-tree conjecture in amended form. A big problem in the area is whether countable graphs are well-quasi-ordered (this implies an abundance of well-quasi ordering conjectures in finite graph theory as well).