Global rigidity of random graphs in $\mathbb{R}$

Richard Montgomery,Rajko Nenadov,Tibor Szabó

2024-01-20

Abstract:Consider the Erdős-Rényi random graph process $\{G_m\}_{m \ge 0}$ in which we start with an empty graph $G_0$ on the vertex set $[n]$, and in each step form $G_i$ from $G_{i-1}$ by adding one new edge chosen uniformly at random. Resolving a conjecture by Benjamini and Tzalik, we give a simple proof that w.h.p. as soon as $G_m$ has minimum degree $2$ it is globally rigid in the following sense: For any function $d \colon E(G_m) \rightarrow \mathbb{R}$, there exists at most one injective function $f \colon [n] \rightarrow \mathbb{R}$ (up to isometry) such that $d(ij) = |f(i) - f(j)|$ for every $ij \in E(G_m)$. We also resolve a related question of Girão, Illingworth, Michel, Powierski, and Scott in the sparse regime for the random graph and give some open problems.

Combinatorics

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is about the global rigidity of random graphs. Specifically, the authors are concerned with whether, in the Erdős - Rényi random graph process, when the minimum degree of the graph \(G_m\) reaches 2, it can be ensured with high probability (w.h.p.) that the graph is globally rigid in the real space \(\mathbb{R}\). Global rigidity means that for any distance function \(d: E(G_m)\to\mathbb{R}\), there exists at most one injective function \(f: [n]\to\mathbb{R}\) (up to isometries) such that \(d(ij) = |f(i)-f(j)|\) holds for every edge \(ij\in E(G_m)\).

The paper mentions that Benjamini and Tzalik proposed a conjecture that when the minimum degree of \(G_m\) reaches 2, the graph should have global rigidity. The authors of this paper confirm this conjecture by providing a simple proof. In addition, they also solve a related problem about sparse random graphs proposed by Girão et al. and raise some open problems.

In summary, the core problem of this paper is to verify and prove the global rigidity of random graphs under specific conditions, and to explore the applications and limitations of this property in different scenarios.