Outliers in the single ring theorem (Q292127)

This article is about spiked models of non-Hermitian random matrices. More specifically, the authors consider matrices of the type \(A+P\), where the rank of \(P\) stays bounded as the dimension goes to infinity and where the matrix \(A\) is a non-Hermitian random matrix, satisfying an isotropy hypothesis: its distribution is invariant under the left and right actions of the unitary group. The macroscopic eigenvalue distribution of such matrices is governed by the so-called single ring theorem, due to \textit{A. Guionnet} et al. [Ann. Math. (2) 174, No. 2, 1189--1217 (2011; Zbl 1239.15018)]. At first it is shown that if \(P\) has some eigenvalues out of the maximal circle of the single ring, then \(A + P\) has some eigenvalues (called outliers) in the neighborhood of those of \(P\), which is not the case for the eigenvalues of \(P\) in the inner cycle of the single ring. Then, the fluctuations of the outliers of \(A\) are studied around the eigenvalues of \(P\) and it is shown that they are distributed as the eigenvalues of some finite-dimensional random matrices. Such a kind of fluctuations had already been shown for Hermitian models. More surprising facts are that outliers can here have very various rates of convergence to their limits (depending on the Jordan canonical form of \(P\)) and that some correlations can appear between outliers at a macroscopic distance from each other (a fact already noticed by \textit{A. Knowles} and \textit{J. Yin} [Ann. Probab. 42, No. 5, 1980--2031 (2014; Zbl 1306.15034)] in the Hermitian case, but only for non-Gaussian models, whereas spiked Gaussian matrices belong to the model described and can have such correlated outliers). The first result generalizes a result by \textit{T. Tao} [Probab. Theory Relat. Fields 155, No. 1--2, 231--263 (2013; Zbl 1261.60009); erratum ibid. 157, No. 1--2, 511--514 (2013; Zbl 1277.60018)] proved specifically for matrices with i.i.d. entries, whereas the second one (about the fluctuations) is new.