The Setting
Consider the conditional distribution of spins on an observed set as the prescribed spins on a source set vary, with fixed signs on a pinned set . The source prescription may also be sampled from a probability distribution.
A finite ferromagnetic weighted multigraph has a finite vertex set , a finite set of separately indexed edges, endpoint maps with , and real couplings .✦ Parallel edges are allowed. A spin configuration assigns a physical spin to every vertex.✦ Two configurations agree on a set when their restrictions to coincide.✦
For a real field , the Ising interaction exponent and the Boltzmann weight are (1) The constrained partition sum for a condition on configurations is (2)✦
Fix pairwise disjoint sets and configurations . The unnormalized joint marginal with pins is the constrained sum (3)✦ where subscripts denote restriction and a list of constraints means their intersection.
The Gibbs measure normalizes the weights in (1). The conditional spin probability is the probability of the observed prescription , given the source prescription and the pins , expressed through the constrained sums of (2): (4)✦
The extreme log odds is the following combination of the joint marginals (3), with and the constant positive and negative configurations: (5)✦
For real mixing weights , the mixture spin probability is the weighted sum of the conditional probabilities (4): (6)✦
The zero-field Fortuin-Kasteleyn terminal connection probability is the probability that the two terminals are connected in the random-cluster model with cluster weight and edge probabilities on the graph obtained from by deleting and its incident edges and identifying and separately as the two terminals. Parallel edges remain distinct; loops created by identification contribute constants and cancel from terminal connection probabilities.
The Claim
Theorem 1. Let be a finite ferromagnetic weighted multigraph with real field , let be pairwise disjoint, and fix pin signs . With the extreme log odds of (5), every observed prescription and every pair of source prescriptions satisfy (7)✦ Let be arbitrary probability weights, so that . For the mixture probabilities of (6), (8)✦
In the paper, the result reads: Theorem 1.1 bounds the ratio of actual normalized configuration probabilities with arbitrary probability mixtures on both sides by and , with intermediate bounds and and the extreme log odds estimate , while mixing preserves the same constants without another factor in the exponent.
What Is Checked
The claims (7) and (8) are stated over finite ferromagnetic weighted multigraphs, arbitrary finite real fields, pairwise disjoint observed, source and pinned sets, and arbitrary spin prescriptions; the mixture claim additionally requires two nonnegative weight families of total mass one.