English

Characterizing graphs with fully positive semidefinite $Q$-matrices

Combinatorics 2023-05-09 v2

Abstract

For qRq\in\mathbb{R}, the QQ-matrix Q=QqQ=Q_q of a connected simple graph G=(V,E)G=(V,E) is Qq=(q(x,y))x,yVQ_q=(q^{\partial(x,y)})_{x,y\in V}, where \partial denotes the path-length distance. Describing the set π(G)\pi(G) consisting of those qRq\in \mathbb{R} for which QqQ_q is positive semidefinite is fundamental in asymptotic spectral analysis of graphs from the viewpoint of quantum probability theory. Assume that GG has at least two vertices. Then π(G)\pi(G) is easily seen to be a nonempty closed subset of the interval [1,1][-1,1]. In this note, we show that π(G)=[1,1]\pi(G)=[-1,1] if and only if GG is isometrically embeddable into a hypercube (infinite-dimensional if GG is infinite) if and only if GG is bipartite and does not possess certain five-vertex configurations, an example of which is an induced K2,3K_{2,3}.

Keywords

Cite

@article{arxiv.2208.11002,
  title  = {Characterizing graphs with fully positive semidefinite $Q$-matrices},
  author = {Hajime Tanaka},
  journal= {arXiv preprint arXiv:2208.11002},
  year   = {2023}
}

Comments

6 pages

R2 v1 2026-06-25T01:54:21.823Z