Airy edge universality for general invariant ensemble additions

From papers

Fix β>0\beta>0 and deterministic sequences of NN-tuples Vecai(N)Vec{a}_i(N), for i=1,2,,ki=1,2,\ldots,k, and let

Vecλ(N)=Veca1(N)++Vecak(N).Vec{\lambda}(N)=Vec{a}_1(N)+\cdots+Vec{a}_k(N).

Assume that, for each ii, the empirical measure of Vecai(N)Vec{a}_i(N) converges as NN\to\infty to a deterministic probability measure muimu_i on R\mathbb{R}. Let boxplusboxplus denote free convolution and set

mu=mu1\boxplusmu2\boxplusmuk.mu=mu_1\boxplusmu_2\boxplus\cdots\boxplusmu_k.

Assume without loss of generality that μ+μ|\mu_+|\geq|\mu_-|.

Airy edge universality conjecture. Under some mild conditions on the summands, the upper edge limit of Vecλ(N)Vec{\lambda}(N) near mu+mu_+ is still given by Airy(β)\operatorname{Airy}(\beta) under proper rescaling.

This conjecture proposes universality of the upper-edge limit beyond the Gaussian and Laguerre β\beta-ensemble additions treated in the paper, suggesting that the limiting edge behavior does not depend on the individual eigenvalue distributions of the summands. The precise mild conditions remain unspecified in the source.

Progress summary

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Sources & referencesView supporting material

Primary source

David Keating and Jiaming Xu, “Airy limit for β-additions through Dunkl operators”, arXiv:2411.12149 (2026).

Solutions 0

No solutions have been posted yet.