English

Hidden temperature in the KMP model

Probability 2024-06-04 v2 Statistical Mechanics

Abstract

In the Kipnis Marchioro Presutti (KMP) model a positive energy ζi\zeta_i is associated with each vertex ii of a finite graph with a boundary. When a Poisson clock rings at an edge ijij with energies ζi,ζj\zeta_i,\zeta_j, those values are substituted by U(ζi+ζj)U(\zeta_i+\zeta_j) and (1U)(ζi+ζj)(1-U)(\zeta_i+\zeta_j), respectively, where UU is a uniform random variable in (0,1)(0,1). A value Tj0T_j\ge 0 is fixed at each boundary vertex jj. The dynamics is defined in such way that the resulting Markov process ζ(t)\zeta(t), satisfies that ζj(t)\zeta_j(t) is exponential with mean TjT_j, for each boundary vertex jj, for all tt. We show that the invariant measure is the distribution of a vector ζ\zeta with coordinates ζi=TiXi\zeta_i=T_i X_i, where XiX_i are iid exponential(1)(1) random variables, the law of TT is the invariant measure for an opinion random averaging/gossip model with the same boundary conditions of ζ\zeta, and the vectors XX and TT are independent. The result confirms a conjecture based on the large deviations of the model. When the graph is one-dimensional, we bound the correlations of the invariant measure and perform the hydrostatic limit. We show that the empirical measure of a configuration chosen with the invariant measure converges to the linear interpolation of the boundary values.

Keywords

Cite

@article{arxiv.2310.01672,
  title  = {Hidden temperature in the KMP model},
  author = {Anna De Masi and Pablo A. Ferrari and Davide Gabrielli},
  journal= {arXiv preprint arXiv:2310.01672},
  year   = {2024}
}

Comments

35 pages, 4 figures

R2 v1 2026-06-28T12:38:56.581Z