On the probability of planarity of a random graph near the critical point

TL;DR

This paper precisely determines the probability that a random graph near the critical point is planar, revealing it approaches approximately 0.998, and extends the analysis to minor-closed graph classes.

Contribution

It provides an exact analytic expression for the planarity probability of random graphs near the critical point, advancing understanding of phase transitions in random graph properties.

Findings

Probability of planarity approaches 0.99780 at the critical point

Probability of being series-parallel converges to 0.98003

Derived explicit formulas for planarity near the critical point

Abstract

Consider the uniform random graph $G(n,M)$ with $n$ vertices and $M$ edges. Erd\H{o}s and R\'enyi (1960) conjectured that the limit $$ \lim_{n \to \infty} \Pr\{G(n,\textstyle{n\over 2}) is planar}} $$ exists and is a constant strictly between 0 and 1. \L uczak, Pittel and Wierman (1994) proved this conjecture and Janson, \L uczak, Knuth and Pittel (1993) gave lower and upper bounds for this probability. In this paper we determine the exact probability of a random graph being planar near the critical point $M=n/2$. For each $\lambda$, we find an exact analytic expression for $$ p(\lambda) = \lim_{n \to \infty} \Pr{G(n,\textstyle{n\over 2}(1+\lambda n^{-1/3})) is planar}.$$ In particular, we obtain $p(0) \approx 0.99780$. We extend these results to classes of graphs closed under taking minors. As an example, we show that the probability of $G(n,\textstyle{n\over 2})$ being series-parallel converges to 0.98003. For the sake of completeness and exposition we reprove in a concise way several basic properties we need of a random graph near the critical point.