Solution to a problem of Katona on counting cliques of weighted graphs

preprint2023arXivOpen access

Abstract

A subset II of the vertex set V(G)V(G) of a graph GG is called a kk-clique independent set of GG if no kk vertices in II form a kk-clique of GG. An independent set is a 22-clique independent set. Let πk(G)π_k(G) denote the number of kk-cliques of GG. For a function w:V(G)→{0,1,2,… }w: V(G) \rightarrow \{0, 1, 2, \dots\}, let G(w)G(w) be the graph obtained from GG by replacing each vertex vv by a w(v)w(v)-clique KvK^v and making each vertex of KuK^u adjacent to each vertex of KvK^v for each edge {u,v}\{u,v\} of GG. For an integer m≥1m \geq 1, consider any ww with ∑v∈V(G)w(v)=m\sum_{v \in V(G)} w(v) = m. For U⊆V(G)U \subseteq V(G), we say that ww is uniform on UU if w(v)=0w(v) = 0 for each v∈V(G)∖Uv \in V(G) \setminus U and, for each u∈Uu \in U, w(u)=⌊m/∣U∣⌋w(u) = \left\lfloor m/|U| \right\rfloor or w(u)=⌈m/∣U∣⌉w(u) = \left\lceil m/|U| \right\rceil. Katona asked if πk(G(w))π_k(G(w)) is smallest when ww is uniform on a largest kk-clique independent set of GG. He placed particular emphasis on the Sperner graph BnB_n, given by V(Bn)={X ⁣:X⊆{1,…,n}}V(B_n) = \{X \colon X \subseteq \{1, \dots, n\}\} and E(Bn)={{X,Y} ⁣:X⊊Y∈V(Bn)}E(B_n) = \{\{X,Y\} \colon X \subsetneq Y \in V(B_n)\}. He provided an affirmative answer for k=2k = 2 (and any GG). We determine graphs for which the answer is negative for every k≥3k \geq 3. These include BnB_n for n≥2n \geq 2. Generalizing Sperner's Theorem and a recent result of Qian, Engel and Xu, we show that πk(Bn(w))π_k(B_n(w)) is smallest when ww is uniform on a largest independent set of BnB_n. We also show that the same holds for complete multipartite graphs and chordal graphs. We show that this is not true of every graph, using a deep result of Bohman on triangle-free graphs.

Explore this paper’s authors, topics and related work.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Reviews 0

Write a reviewWrite

No reviews yet.

Discussion 0

Add a commentWrite

No comments yet.