arXiv subjects

Manfred Scheucher

Publications and source records attributed to Manfred Scheucher.

At least 19 recordsLinked to original sources

arXiv2025

Investigating Simple Drawings of $K_n$ using SAT

We present a SAT framework which allows to investigate properties of simple drawings of the complete graph $K_n$ using the power of AI. In contrast to classic imperative programming, where a program is operated step by step, our framework models mathematical questions as Boolean formulas which are then solved using modern SAT solvers. Our framework for simple drawings is based on a characterization via rotation systems and finite forbidden substructures. We showcase its universality by addressing various open problems, reproving previous computational results and deriving several new computational results. In particular, we test and progress on several unavoidable configurations such as variants of Rafla's conjecture on plane Hamiltonian cycles, Harborth's conjecture on empty triangles, and crossing families for general simple drawings as well as for various subclasses. Moreover, based our computational results we propose some new challenging conjectures.

cs.CG↗

arXiv2025

On Triangular Separation of Bichromatic Point Sets

We address the problem of computing the minimum number of triangles to separate a set of blue points from a set of red points in $\mathbb{R}^2$. A set of triangles is a \emph{separator} of one color from the other if every point of that color is contained in some triangle and no triangle contains points of both colors. We consider several variants of the problem depending on whether the triangles are allowed to overlap or not and whether all points or just the blue points need to be contained in a triangle. We show that computing the minimum cardinality triangular separator of a set of blue points from a set of red points is NP-hard and further investigate worst case bounds on the minimum cardinality of triangular separators for a bichromatic set of $n$ points.

cs.CG↗

arXiv2024

Subgraph-universal planar graphs for trees

We show that there exists an outerplanar graph on $O(n^{c})$ vertices for $c = \log_2(3+\sqrt{10}) \approx 2.623$ that contains every tree on $n$ vertices as a subgraph. This extends a result of Chung and Graham from 1983 who showed that there exist (non-planar) $n$-vertex graphs with $O(n \log n)$ edges that contain all trees on $n$ vertices as subgraphs and a result from Gol'dberg and Livshits from 1968 who showed that there exists a universal tree for $n$-vertex trees on $n^{O(\log(n))}$ vertices. Furthermore, we determine the number of vertices needed in the worst case for a planar graph to contain three given trees as subgraph to be on the order of $\frac{3}{2}n$, even if the three trees are caterpillars. This answers a question recently posed by Alecu et al. in 2024. Lastly, we investigate (outer)planar graphs containing all (outer)planar graphs as subgraph, determining exponential lower bounds in both cases. We also construct a planar graph on $n^{O(\log(n))}$ vertices containing all $n$-vertex outerplanar graphs as subgraphs.

math.CO↗

arXiv2024

Holes in Convex and Simple Drawings

Gons and holes in point sets have been extensively studied in the literature. For simple drawings of the complete graph a generalization of the Erd\H{o}s--Szekeres theorem is known and empty triangles have been investigated. We introduce a notion of $k$-holes for simple drawings and survey generalizations thereof, like empty $k$-cycles. We present a family of simple drawings without $4$-holes and prove a generalization of Gerken's empty hexagon theorem for convex drawings. A crucial intermediate step is the structural investigation of pseudolinear subdrawings in convex drawings. With respect to empty $k$-cycles, we show the existence of empty $4$-cycles in every simple drawing of $K_n$ and give a construction that admits only $\Theta(n^2)$ of them.

cs.CG↗

arXiv2024

Plane Hamiltonian Cycles in Convex Drawings

A conjecture by Rafla from 1988 asserts that every simple drawing of the complete graph $K_n$ admits a plane Hamiltonian cycle. It turned out that already the existence of much simpler non-crossing substructures in such drawings is hard to prove. Recent progress was made by Aichholzer et al. and by Suk and Zeng who proved the existence of a plane path of length $\Omega(\log n / \log \log n)$ and of a plane matching of size $\Omega(n^{1/2})$ in every simple drawing of $K_n$. Instead of studying simpler substructures, we prove Rafla's conjecture for the subclass of convex drawings, the most general class in the convexity hierarchy introduced by Arroyo et al. Moreover, we show that every convex drawing of $K_n$ contains a plane Hamiltonian path between each pair of vertices (Hamiltonian connectivity) and a plane $k$-cycle for each $3 \leq k \leq n$ (pancyclicity), and present further results on maximal plane subdrawings.

cs.CG↗

arXiv2024

Happy Ending: An Empty Hexagon in Every Set of 30 Points

Satisfiability solving has been used to tackle a range of long-standing open math problems in recent years. We add another success by solving a geometry problem that originated a century ago. In the 1930s, Esther Klein's exploration of unavoidable shapes in planar point sets in general position showed that every set of five points includes four points in convex position. For a long time, it was open if an empty hexagon, i.e., six points in convex position without a point inside, can be avoided. In 2006, Gerken and Nicol\'as independently proved that the answer is no. We establish the exact bound: Every 30-point set in the plane in general position contains an empty hexagon. Our key contributions include an effective, compact encoding and a search-space partitioning strategy enabling linear-time speedups even when using thousands of cores.

cs.CG↗

arXiv2024

An Improved Lower Bound on the Number of Pseudoline Arrangements

Arrangements of pseudolines are classic objects in discrete and computational geometry. They have been studied with increasing intensity since their introduction almost 100 years ago. The study of the number $B_n$ of non-isomorphic simple arrangements of $n$ pseudolines goes back to Goodman and Pollack, Knuth, and others. It is known that $B_n$ is in the order of $2^{\Theta(n^2)}$ and finding asymptotic bounds on $b_n = \frac{\log_2(B_n)}{n^2}$ remains a challenging task. In 2011, Felsner and Valtr showed that $0.1887 \leq b_n \le 0.6571$ for sufficiently large $n$. The upper bound remains untouched but in 2020 Dumitrescu and Mandal improved the lower bound constant to $0.2083$. Their approach utilizes the known values of $B_n$ for up to $n=12$. We tackle the lower bound by utilizing dynamic programming and the Lindstr\"om-Gessel-Viennot lemma. Our new bound is $b_n \geq 0.2721$ for sufficiently large $n$. The result is based on a delicate interplay of theoretical ideas and computer assistance.

math.CO↗

arXiv2024

Finding hardness reductions automatically using SAT solvers

In this article, we show that the completion problem, i.e. the decision problem whether a partial structure can be completed to a full structure, is NP-complete for many combinatorial structures. While the gadgets for most reductions in literature are found by hand, we present an algorithm to construct gadgets in a fully automated way. Using our framework which is based on SAT, we present the first thorough study of the completion problem on sign mappings with forbidden substructures by classifying thousands of structures for which the completion problem is NP-complete. Our list in particular includes interior triple systems, which were introduced by Knuth towards an axiomatization of planar point configurations. Last but not least, we give an infinite family of structures generalizing interior triple system to higher dimensions for which the completion problem is NP-complete.

cs.CC↗

arXiv2023

Saturation results around the Erd\H{o}s--Szekeres problem

In this paper, we consider saturation problems related to the celebrated Erd\H{o}s--Szekeres convex polygon problem. For each $n \ge 7$, we construct a planar point set of size $(7/8) \cdot 2^{n-2}$ which is saturated for convex $n$-gons. That is, the set contains no $n$ points in convex position while the addition of any new point creates such a configuration. This demonstrates that the saturation number is smaller than the Ramsey number for the Erd\H{o}s--Szekeres problem. The proof also shows that the original Erd\H{o}s--Szekeres construction is indeed saturated. Our construction is based on a similar improvement for the saturation version of the cups-versus-caps theorem. Moreover, we consider the generalization of the cups-versus-caps theorem to monotone paths in ordered hypergraphs. In contrast to the geometric setting, we show that this abstract saturation number is always equal to the corresponding Ramsey number.

math.CO↗

arXiv2023

Flip Graph Connectivity for Arrangements of Pseudolines and Pseudocircles

Flip graphs of combinatorial and geometric objects are at the heart of many deep structural insights and connections between different branches of discrete mathematics and computer science. They also provide a natural framework for the study of reconfiguration problems. We study flip graphs of arrangements of pseudolines and of arrangements of pseudocircles, which are combinatorial generalizations of lines and circles, respectively. In both cases we consider triangle flips as local transformation and prove conjectures regarding their connectivity. In the case of $n$ pseudolines we show that the connectivity of the flip graph equals its minimum degree, which is exactly $n-2$. For the proof we introduce the class of shellable line arrangements, which serve as reference objects for the construction of disjoint paths. In fact, shellable arrangements are elements of a flip graph of line arrangements which are vertices of a polytope (Felsner and Ziegler; DM 241 (2001), 301--312). This polytope forms a cluster of good connectivity in the flip graph of pseudolines. In the case of pseudocircles we show that triangle flips induce a connected flip graph on \emph{intersecting} arrangements and also on cylindrical intersecting arrangements. The result for cylindrical arrangements is used in the proof for intersecting arrangements. We also show that in both settings the diameter of the flip graph is in $\Theta(n^3)$. Our constructions make essential use of variants of the sweeping lemma for pseudocircle arrangements (Snoeyink and Hershberger; Proc.\ SoCG 1989: 354--363). We finally study cylindrical arrangements in their own right and provide new combinatorial characterizations of this class.

math.CO↗

arXiv2023

Bichromatic Perfect Matchings with Crossings

We consider bichromatic point sets with $n$ red and $n$ blue points and study straight-line bichromatic perfect matchings on them. We show that every such point set in convex position admits a matching with at least $\frac{3n^2}{8}-\frac{n}{2}+c$ crossings, for some $ -\frac{1}{2} \leq c \leq \frac{1}{8}$. This bound is tight since for any $k> \frac{3n^2}{8} -\frac{n}{2}+\frac{1}{8}$ there exist bichromatic point sets that do not admit any perfect matching with $k$ crossings.

cs.CG↗

arXiv2023

Using SAT to study plane Hamiltonian substructures in simple drawings

In 1988 Rafla conjectured that every simple drawing of a complete graph $K_n$ contains a plane, i.e., non-crossing, Hamiltonian cycle. The conjecture is far from being resolved. The lower bounds for plane paths and plane matchings have recently been raised to $(\log n)^{1-o(1)}$ and $\Omega(\sqrt{n})$, respectively. We develop a SAT framework which allows the study of simple drawings of $K_n$. Based on the computational data we conjecture that every simple drawing of $K_n$ contains a plane Hamiltonian subgraph with $2n-3$ edges. We prove this strengthening of Rafla's conjecture for convex drawings, a rich subclass of simple drawings. Our computer experiments also led to other new challenging conjectures regarding plane substructures in simple drawings of complete graphs.

cs.CG↗

arXiv2023

Roudneff's Conjecture in Dimension $4$

J.-P. Roudneff conjectured in 1991 that every arrangement of $n \ge 2d+1\ge 5$ pseudohyperplanes in the real projective space $\mathbb{P}^d$ has at most $\sum_{i=0}^{d-2} \binom{n-1}{i}$ complete cells (i.e., cells bounded by each hyperplane). The conjecture is true for $d=2,3$ and for arrangements arising from Lawrence oriented matroids. The main result of this manuscript is to show the validity of Roudneff's conjecture for $d=4$. Moreover, based on computational data we conjecture that the maximum number of complete cells is only obtained by cyclic arrangements.

math.CO↗

arXiv2023

An extension theorem for signotopes

In 1926, Levi showed that, for every pseudoline arrangement $\mathcal{A}$ and two points in the plane, $\mathcal{A}$ can be extended by a pseudoline which contains the two prescribed points. Later extendability was studied for arrangements of pseudohyperplanes in higher dimensions. While the extendability of an arrangement of proper hyperplanes in $\mathbb{R}^d$ with a hyperplane containing $d$ prescribed points is trivial, Richter-Gebert found an arrangement of pseudoplanes in $\mathbb{R}^3$ which cannot be extended with a pseudoplane containing two particular prescribed points. In this article, we investigate the extendability of signotopes, which are a combinatorial structure encoding a rich subclass of pseudohyperplane arrangements. Our main result is that signotopes of odd rank are extendable in the sense that for two prescribed crossing points we can add an element containing them. Moreover, we conjecture that in all even ranks $r \geq 4$ there exist signotopes which are not extendable for two prescribed points. Our conjecture is supported by examples in ranks 4, 6, 8, 10, and 12 that were found with a SAT based approach.

math.CO↗

arXiv2022

Blocking Delaunay Triangulations from the Exterior

Given two distinct point sets $P$ and $Q$ in the plane, we say that $Q$ \emph{blocks} $P$ if no two points of $P$ are adjacent in any Delaunay triangulation of $P\cup Q$. Aichholzer et al. (2013) showed that any set $P$ of $n$ points in general position can be blocked by $\frac{3}{2}n$ points and that every set $P$ of $n$ points in convex position can be blocked by $\frac{5}{4}n$ points. Moreover, they conjectured that, if $P$ is in convex position, $n$ blocking points are sufficient and necessary. The necessity was recently shown by Biniaz (2021) who proved that every point set in general position requires $n$ blocking points. Here we investigate the variant, where blocking points can only lie outside of the convex hull of the given point set. We show that $\frac{5}{4}n-O(1)$ such \emph{exterior-blocking} points are sometimes necessary, even if the given point set is in convex position. As a consequence we obtain that, if the conjecture of Aichholzer et al. for the original setting was true, then minimal blocking sets of some point configurations $P$ would have to contain points inside of the convex hull of $P$.

cs.CG↗

arXiv2022

Arrangements of Pseudocircles: On Digons and Triangles

In this article, we study the cell-structure of simple arrangements of pairwise intersecting pseudocircles. The focus will be on two problems of Gr\"unbaum (1972). First, we discuss the maximum number of digons or touching points. Gr\"unbaum conjectured that there are at most $2n - 2$ digon cells or equivalently at most $2n - 2$ touchings. Agarwal et al. (2004) verified the conjecture for cylindrical arrangements. We show that the conjecture holds for any arrangement which contains three pairwise touching pseudocircles. The proof makes use of the result for cylindrical arrangements. Moreover, we construct non-cylindrical arrangements which attain the maximum of $2n - 2$ touchings and have no triple of pairwise touching pseudocircles. Second, we discuss the minimum number of triangular cells (triangles) in arrangements without digons and touchings. Gr\"unbaum conjectured that such arrangements have $2n - 4$ triangles. Snoeyink and Hershberger (1991) established a lower bound of $\lceil \frac{4}{3}n \rceil$. Felsner and Scheucher (2017) disproved the conjecture and constructed a family of arrangements with only $\lceil \frac{16}{11}n \rceil$ triangles. We provide a construction which shows that $\lceil \frac{4}{3}n \rceil$ is the correct value.

cs.CG↗

arXiv2022

Coloring circle arrangements: New $4$-chromatic planar graphs

Felsner, Hurtado, Noy and Streinu (2000) conjectured that arrangement graphs of simple great-circle arrangements have chromatic number at most $3$. Motivated by this conjecture, we study the colorability of arrangement graphs for different classes of arrangements of (pseudo-)circles. In this paper the conjecture is verified for $\triangle$-saturated pseudocircle arrangements, i.e., for arrangements where one color class of the 2-coloring of faces consists of triangles only, as well as for further classes of (pseudo-)circle arrangements. These results are complemented by a construction which maps $\triangle$-saturated arrangements with a pentagonal face to arrangements with 4-chromatic 4-regular arrangement graphs. This "corona" construction has similarities with the crowning construction introduced by Koester (1985). Based on exhaustive experiments with small arrangements we propose three strengthenings of the original conjecture. We also investigate fractional colorings. It is shown that the arrangement graph of every arrangement $\mathcal{A}$ of pairwise intersecting pseudocircles is "close" to being $3$-colorable. More precisely, the fractional chromatic number $\chi_f(\mathcal{A})$ of the arrangement graph is bounded from above by $\chi_f(\mathcal{A}) \le 3+O(\frac{1}{n})$, where $n$ is the number of pseudocircles of $\mathcal{A}$. Furthermore, we construct an infinite family of $4$-edge-critical $4$-regular planar graphs which are fractionally $3$-colorable. This disproves a conjecture of Gimbel, K\"{u}ndgen, Li, and Thomassen (2019).

math.CO↗

arXiv2022

Erd\H{o}s--Szekeres-type problems in the real projective plane

We consider point sets in the real projective plane $\mathbb{R}P^2$ and explore variants of classical extremal problems about planar point sets in this setting, with a main focus on Erd\H{o}s--Szekeres-type problems. We provide asymptotically tight bounds for a variant of the Erd\H{o}s--Szekeres theorem about point sets in convex position in $\mathbb{R}P^2$, which was initiated by Harborth and M\"oller in 1994. The notion of convex position in $\mathbb{R}P^2$ agrees with the definition of convex sets introduced by Steinitz in 1913. For $k \geq 3$, an (\affine) $k$-hole in a finite set $S \subseteq \mathbb{R}^2$ is a set of $k$ points from $S$ in convex position with no point of $S$ in the interior of their convex hull. After introducing a new notion of $k$-holes for points sets from $\mathbb{R}P^2$, called projective $k$-holes, we find arbitrarily large finite sets of points from $\mathbb{R}P^2$ with no \projective 8-holes, providing an analogue of a classical planar construction by Horton from 1983. We also prove that they contain only quadratically many \projective $k$-holes for $k \leq 7$. On the other hand, we show that the number of $k$-holes can be substantially larger in~$\mathbb{R}P^2$ than in $\mathbb{R}^2$ by constructing, for every $k \in \{3,\dots,6\}$, sets of $n$ points from $\mathbb{R}^2 \subset \mathbb{R}P^2$ with $\Omega(n^{3-3/5k})$ \projective $k$-holes and only $O(n^2)$ \affine $k$-holes. Last but not least, we prove several other results, for example about projective holes in random point sets in $\mathbb{R}P^2$ and about some algorithmic aspects. The study of extremal problems about point sets in $\mathbb{R}P^2$ opens a new area of research, which we support by posing several open problems.

math.CO↗