On Spatial Conditioning of the Spectrum of Discrete Random Schrödinger Operators
arXiv:2101.00319
Abstract
Consider a random Schrödinger-type operator of the form acting on a general graph , where is the generator of a Markov process on , is a deterministic potential with sufficient growth (so that has a purely discrete spectrum), and is a random noise with at-most-exponential tails. We prove that 's eigenvalue point process is number rigid in the sense of Ghosh and Peres (Duke Math. J. 166 (2017), no. 10, 1789--1858); that is, the number of eigenvalues in any bounded domain is determined by the configuration of eigenvalues outside of . Our general setting allows to treat cases where could be non-symmetric (hence is non-self-adjoint) and has long-range dependence. Our strategy of proof consists of controlling the variance of the trace of the semigroup using the Feynman-Kac formula.
arXiv comment32 pages; revised version incorporating referee comments and fixing a few misprints and errors. Accepted in the Journal of Spectral Theory
References in corpus (6)
- Beta ensembles, stochastic Airy spectrum, and a diffusion
- From Random Matrices to Stochastic Operators
- Rigidity of Determinantal Point Processes with the Airy, the Bessel and the Gamma Kernel
- Generalized stealthy hyperuniform processes : maximal rigidity and the bounded holes conjecture
- Rigidity and tolerance for perturbed lattices
- Rigidity hierarchy in random point fields: random polynomials and determinantal processes