Large gaps between random eigenvalues (Q984450)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Large gaps between random eigenvalues |
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Large gaps between random eigenvalues |
scientific article |
Statements
Large gaps between random eigenvalues (English)
0 references
19 July 2010
0 references
One of the classical topics in random matrix theory is the local behavior of eigenvalues, modeling the energy levels of a physical system. Wigner put forward a prediction for the asymptotic probability of having no GOE eigenvalues in a fixed interval of the real line, as matrix size tends to infinity and the spectrum is properly rescaled. \textit{F. J. Dyson} [J. Math. Phys. 3, 140--156, 157--165, 166--175 (1962; Zbl 0105.41604)] stated a corrected version for all \(\beta\)-ensembles, and a rigorous proof for the classical Wigner-Dyson cases \(\beta = 1, 2, 4\) was subsequently given by several authors, e.g., \textit{P. A. Deift, A. R. Its} and \textit{X. Zhou} [Ann. Math. (2) 146, No. 1, 149--235 (1997; Zbl 0936.47028)]. In the paper under review, the authors provide a proof for general \(\beta\). It is based on their recent work [Invent. Math. 177, No. 3, 463--508 (2009; Zbl 1204.60012)] which represents the point process limit of the bulk eigenvalues in terms of the Brownian carousel, a deterministic function of the Brownian motion on the Poincaré disk. This makes it possible to express the gap probability of interest in terms of a passage probability for a related process and invoke a version of the Cameron-Martin-Girsanov formula.
0 references
random matrices
0 references
\(\beta\)-ensembles
0 references
large deviations
0 references
Brownian carousel
0 references
0.8162633180618286
0 references
0.8003583550453186
0 references
0.7944035530090332
0 references
0.7933517694473267
0 references
0.7878544330596924
0 references