Finitary codings for spatial mixing Markov random fields

Measure-theoretic ergodic theory (28D99) Dynamical aspects of statistical mechanics (37A60) Stationary stochastic processes (60G10) Markov processes (60J99) Interacting random processes; statistical mechanics type models; percolation theory (60K35) Lattice systems (Ising, dimer, Potts, etc.) and systems on graphs arising in equilibrium statistical mechanics (82B20)

Abstract: It has been shown by van den Berg and Steif that the sub-critical and critical Ising model on

m a t h b b Z^{d}

is a finitary factor of an i.i.d. process (ffiid), whereas the super-critical model is not. In fact, they showed that the latter is a general phenomenon in that a phase transition presents an obstruction for being ffiid. The question remained whether this is the only such obstruction. We make progress on this, showing that certain spatial mixing conditions (notions of weak dependence on boundary conditions, not to be confused with other notions of mixing in ergodic theory) imply ffiid. Our main result is that weak spatial mixing implies ffiid with power-law tails for the coding radius, and that strong spatial mixing implies ffiid with exponential tails for the coding radius. The weak spatial mixing condition can be relaxed to a condition which is satisfied by some critical two-dimensional models. Using a result of the author, we deduce that strong spatial mixing also implies ffiid with stretched-exponential tails from a finite-valued i.i.d. process. We give several applications to models such as the Potts model, proper colorings, the hard-core model, the Widom--Rowlinson model and the beach model. For instance, for the ferromagnetic

q

-state Potts model on

m a t h b b Z^{d}

at inverse temperature

, we show that it is ffiid with exponential tails if

is sufficiently small, it is ffiid if

, it is not ffiid if

and, when

d = 2

and

, it is ffiid if and only if

q l e 4

.

Review text

The author reminds that the subcritical and critical Ising model on \(\mathbb{Z}^d\) is a finitary factor of an i.i.d. process (ffiid), whereas the super-critical model is not. The paper shows that certain spatial mixing conditions imply ffiid. The main result is that weak spatial mixing implies ffiid with power-law tails for the coding radius, and that strong spatial mixing implies ffiid with exponential tails for the coding radius. The author proves that strong spatial mixing also implies ffiid with stretched-exponential tails from a finite-valued i.i.d. process. The author gives several applications to models such as the Potts model, proper colorings, the hard-core model, the Widom-Rowlinson model and the beach model. Corollary 1.5 summarizes some results. Let \(d \ge 2\) and \(q \ge 2\). There exists \({\beta _0}(d,q) \ge {1 \mathord{\left/{\vphantom {1 {2d}}} \right. \kern-\nulldelimiterspace} {2d}}\) satisfying \({\beta _0}(d,q) = {\beta _c}(d,q)\) if either \(q = 2\) or \(d = 2\), such that the following holds. Let \(\mu \) be a Gibbs measure for the ferromagnetic \(q\)-state Potts model on \(\mathbb{Z}^d\) at inverse temperature \(\beta > 0\). If \(\beta < {\beta _0}(d,q)\) then \(\mu \) is ffiid with exponential tails. If \(\beta < {\beta _c}(d,q)\) then \(\mu \) is ffiid with power-law tails. If \(\beta > {\beta _c}(d,q)\) then \(\mu \) is not ffiid. If \(\beta = {\beta _c}(d,q)\) and \(d = 2\), then \(\mu \) is ffiid if and only if \(q \le 4\).

Recommendations

Cites work

Cited in

(19)

This page was built for publication: Finitary codings for spatial mixing Markov random fields

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q784183)