Edge rigidity and universality of random regular graphs of intermediate degree
Abstract
For random d-regular graphs on N vertices with 1 d N2/3, we develop a d-1/2 expansion of the local eigenvalue distribution about the Kesten-McKay law up to order d-3. This result is valid up to the edge of the spectrum. It implies that the eigenvalues of such random regular graphs are more rigid than those of Erdos-R\'enyi graphs of the same average degree. As a first application, for 1 d N2/3, we show that all nontrivial eigenvalues of the adjacency matrix are with very high probability bounded in absolute value by (2 + o(1)) d - 1. As a second application, for N2/9 d N1/3, we prove that the extremal eigenvalues are concentrated at scale N-2/3 and their fluctuations are governed by Tracy-Widom statistics. Thus, in the same regime of d, 52\% of all d-regular graphs have second-largest eigenvalue strictly less than 2 d - 1.
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.