On the evolution of structure in triangle-free graphs

Abstract

We study the typical structure and the number of triangle-free graphs with n vertices and m edges where m is large enough so that a typical triangle-free graph has a cut containing nearly all of its edges, but may not be bipartite. Erdos, Kleitman, and Rothschild showed that almost every triangle-free graph is bipartite. Osthus, Pr\"omel, and Taraz later showed that for m (1+ε)34n3/2 n, almost every triangle-free graph on n vertices and m edges is bipartite. Here we give a precise characterization of the distribution of edges within each part of the max cut of a uniformly chosen triangle-free graph G on n vertices and m edges, for a larger range of densities with m=(n3/2 n). Using this characterization, we describe the evolution of the structure of typical triangle-free graphs as the density changes. We show that as the number of edges decreases below 34 n3/2 n, the following structural changes occur in G: -Isolated edges, then trees, then more complex subgraphs emerge as `defect edges', edges within parts of a max cut of G. The distribution of defect edges is first that of independent Erdos-R\'enyi random graphs, then that of independent exponential random graphs, conditioned on a small maximum degree and no triangles. -There is a sharp threshold for 3-colorability at m 24 n3/2 n and a sharp threshold between 4-colorability and unbounded chromatic number at m14n3/2 n. -Giant components emerge in the defect edges at m14 n3/2 n. We use these results to prove asymptotic formulas for the number of triangle-free graphs at these densities. We likewise prove analogous results for the random graph G(n,p) conditioned on triangle-freeness.

0

Turn this paper into a full lesson

ArcXiv compiles a staged curriculum from this paper: 8-12 lessons across beginner → advanced, synthesised section guides, visuals, flashcards, a quiz, exercises, and on-demand deep dives per section. Grounded in the abstract, never invented.

Discussion (0)

Sign in to join the discussion.

Loading comments…