Spin-glass stochastic stability: a rigorous proof (Q2572601): Difference between revisions
From MaRDI portal
Created claim: Wikidata QID (P12): Q58295812, #quickstatements; #temporary_batch_1707252663060 |
Changed an Item |
||
(2 intermediate revisions by 2 users not shown) | |||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
Property / OpenAlex ID | |||
Property / OpenAlex ID: W2111372674 / rank | |||
Normal rank | |||
Property / arXiv ID | |||
Property / arXiv ID: math-ph/0408002 / rank | |||
Normal rank |
Latest revision as of 08:14, 19 April 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Spin-glass stochastic stability: a rigorous proof |
scientific article |
Statements
Spin-glass stochastic stability: a rigorous proof (English)
0 references
10 November 2005
0 references
In the context of spin glasses, stochastic stability says that a suitable class of perturbations of the spin glass Hamiltonian (perturbations which, typically, involve a change of temperature of order of \(1/V\), \(V\) being the size of the system) produces very small changes in the quenched equilibrium state and that such changes vanish in the thermodynamic limit. This property implies, in the infinite volume limit, that the average of some particular observables (called ``zero-average quantities'' in this paper) vanishes. For mean field spin glasses, these observables are special polynomials of the usual overlaps \(q_{\sigma,\tau}=1/N\sum_{i\leq N}\sigma_i\tau_i\) between pairs of replica \(\sigma,\tau\). For more general (e.g., finite-range) but still Gaussian spin glass models, the ``zero average quantities'' are functions of a generalized overlap, which is unambiguously identified from the covariance matrix of the Hamiltonian. In this paper, it is shown that stochastic stability holds in \(\beta\)-average, i.e., that the average of the ``zero average quantities'' vanishes like \(1/V\) if integrated on an arbitrarily small but finite interval of temperatures, for a general class of Gaussian spin glass models. Of course, this leaves open the problem of proving that stochastic stability holds for a given inverse temperature \(\beta\).
0 references
stochastic stability
0 references
Gaussian spin glasses
0 references