Random band matrices in the delocalized phase, III: Averaging fluctuations
arXiv:1807.02447
Abstract
We consider a general class of symmetric or Hermitian random band matrices in any dimension , where the entries are independent, centered random variables with variances . We assume that vanishes if exceeds the band width , and we are interested in the mesoscopic scale with . Define the {\it{generalized resolvent}} of as , where is a deterministic diagonal matrix with entries for all . Then we establish a precise high-probability bound on certain averages of polynomials of the resolvent entries. As an application of this fluctuation averaging result, we give a self-contained proof for the delocalization of random band matrices in dimensions . More precisely, for any fixed , we prove that the bulk eigenvectors of are delocalized in certain averaged sense if . This improves the corresponding results in \cite{HeMa2018} under the assumption , and in \cite{ErdKno2013,ErdKno2011} under the assumption . For 1D random band matrices, our fluctuation averaging result was used in \cite{PartII,PartI} to prove the delocalization conjecture and bulk universality for random band matrices with .
arXiv comment65 pages
References in corpus (6)
- Spectral Statistics of Erd{\H o}s-Rényi Graphs II: Eigenvalue Spacing and the Extreme Eigenvalues
- Universality of covariance matrices
- Averaging Fluctuations in Resolvents of Random Band Matrices
- Eigenvector localization for random band matrices with power law band width
- Quantum Diffusion and Delocalization for Band Matrices with General Distribution
- Characteristic polynomials for 1D random band matrices from the localization side
Cited by in corpus (4)
- Anderson-Bernoulli Localization on the 3D lattice and discrete unique continuation principle
- Delocalization and continuous spectrum for ultrametric random operators
- On fluctuations of global and mesoscopic linear eigenvalue statistics of generalized Wigner matrices
- Universality of the least singular value for the sum of random matrices