Hidden temperature in the KMP model

Anna De Masi,Pablo A. Ferrari,Davide Gabrielli
2024-06-03
Abstract:In the Kipnis Marchioro Presutti (KMP) model a positive energy $\zeta_i$ is associated with each vertex $i$ of a finite graph with a boundary. When a Poisson clock rings at an edge $ij$ with energies $\zeta_i,\zeta_j$, those values are substituted by $U(\zeta_i+\zeta_j)$ and $(1-U)(\zeta_i+\zeta_j)$, respectively, where $U$ is a uniform random variable in $(0,1)$. A value $T_j\ge 0$ is fixed at each boundary vertex $j$. The dynamics is defined in such way that the resulting Markov process $\zeta(t)$, satisfies that $\zeta_j(t)$ is exponential with mean $T_j$, for each boundary vertex $j$, for all $t$. We show that the invariant measure is the distribution of a vector $\zeta$ with coordinates $\zeta_i=T_i X_i$, where $X_i$ are iid exponential$(1)$ random variables, the law of $T$ is the invariant measure for an opinion random averaging/gossip model with the same boundary conditions of $\zeta$, and the vectors $X$ and $T$ 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.
Probability,Statistical Mechanics
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to understand the complex statistical structure of stationary non - equilibrium states (SNS) in non - equilibrium statistical mechanics. Specifically, the paper studies the Kipnis - Marchioro - Presutti (KMP) model and introduces hidden variables to better understand the statistical properties of these states. ### Background of the Paper In non - equilibrium situations, the system is in contact with external boundary sources, which have different characteristics, thus creating irreversibility and leading to the emergence of non - equilibrium states. This state is usually accompanied by long - range correlations, and the traditional Gibbs equilibrium framework is no longer applicable. ### Research Objectives The main objectives of the paper are: 1. **Introduce Hidden Variables**: By introducing the hidden variable \(T\), better understand the statistical structure of the stationary non - equilibrium states in the KMP model. 2. **Analyze Invariant Measures**: Study the invariant measures of the KMP model and prove that it can be represented as a mixture of exponential distributions, where the average value is determined by the unique invariant measure of \(T\). 3. **Study the One - Dimensional Case**: On a one - dimensional graph, study the correlations of the invariant measures and the hydrostatic limit. ### Main Results 1. **Properties of Invariant Measures**: - Prove that the KMP process \(\zeta(t)\) preserves the set \(M\), that is, if the initial distribution belongs to \(M\), then the distribution at any time \(t\) also belongs to \(M\). - Prove that the unique invariant measure \(\mu_\zeta\) of the KMP process belongs to \(M\), and the parameter measure is the invariant probability \(\nu_O\) of the opinion model \(O\). 2. **Relationships of Correlations**: - Give the relationship between the correlations of \(\mu_\zeta\) and the correlations of \(\nu_t\). 3. **Results in the One - Dimensional Case**: - Prove that the correlations of the invariant measure of the opinion model on a one - dimensional graph satisfy certain bounds. - Prove that the hydrostatic limit of the empirical measure of the opinion model on a one - dimensional graph converges to the linear interpolation of the boundaries. ### Mathematical Expressions - **Generator Operator of the KMP Process**: \[ L_\zeta f(\zeta):=\sum_{ij\in E}\int_0^1 du\left[f(H_{ij;u}\zeta)-f(\zeta)\right]+\sum_{ij\in\partial E}\int_0^1 du\int_0^\infty db\,e^{-b}\left[f(H_{ij;u,b}\zeta)-f(\zeta)\right], \] where, \[ (H_{ij;u}\zeta)_\ell=\begin{cases} u(\zeta_i+\zeta_j)&\text{if }\ell = i,\\ (1 - u)(\zeta_i+\zeta_j)&\text{if }\ell = j,\\ \zeta_\ell&\text{otherwise}, \end{cases} \] \[ (H_{ij;u,b}\zeta)_\ell=\begin{cases} u(\zeta_i+\zeta_j)&\text{if }\ell = i,\\ bT_j&\text{if }\ell = j\in\partial V,\\ \zeta_\ell&\text{otherwise}. \end{cases} \] - **Generator Operator of the Opinion Model**: \[ L_O f(O):=\sum_{ij\in E}\int_0^1 dv