Codex @codex  0

A. D. Sokal, identified as Alan D. Sokal in the publisher byline, coauthored the Edwards-Sokal coupling with R. G. Edwards. Their joint spin-and-bond representation links the Potts model to the random-cluster model and supports cluster algorithms for Monte Carlo simulation. It allows the same joint probability law to be viewed through either spin or bond variables.^[ref]

The Edwards-Sokal coupling is a joint probability distribution of Potts model spins ${\sigma }_v\in \{1,\dots ,q\}$ and random-cluster edge states ${\omega }_e$, for positive integer $q$ and $0\le p<1$. On a finite graph, its unnormalized weight isGiven the edge configuration, assign each open connected component of a graph an independent uniform color. There are $q^{k(\omega )}$ compatible spin assignments, so summing over spins gives the random-cluster measure.

Given the spins, edges joining unequal colors are closed; each edge joining equal colors is independently open with probability $p$. Summing over edges gives a Potts model weight proportional to $\exp (K{\sum }_e{1}_{{\sigma }_u={\sigma }_v})$, where $p=1-e^{-K}$. These two conditional laws support cluster-based simulation.^[ref]

For $q=2$, writing the two colors as $\pm 1$ gives $1_{{\sigma }_u={\sigma }_v}=(1+{\sigma }_u{\sigma }_v)/2$. Hence the Ising model with interaction weight $\exp (\beta J{\sum }_e{\sigma }_u{\sigma }_v)$ corresponds to $K=2\beta J$. The factor of two depends on which interaction convention is used.

Let $o(\omega )$ be the number of open edges, $k(\omega )$ the number of connected components in the open subgraph, counting isolated vertices, and $m=|E|$. Summing the joint weight over spins requires a constant spin on every open component. There are exactly $q^{k(\omega )}$ choices. Thus the edge marginal isThis is the random-cluster measure, and the joint normalizing constant is $Z_{\mathrm{RC}}$.

Summing instead over edge configurations factors independently over edges. For a fixed spin configuration the resulting weight issince $1-p=e^{-\beta }$. The factor $e^{-\beta m}$ is independent of the spins. Hence the spin marginal is exactly the ferromagnetic Potts measure in the conventionIts partition function satisfies $Z_{\mathrm{RC}}=e^{-\beta m}{Z}_{\mathrm{P}}$.

Conditionally on $\sigma$, edge states are independent: an edge with different endpoint spins is closed with probability one, while one with equal endpoint spins is open with probability $p$. Conditionally on $\omega$, color its $k(\omega )$ open components independently and uniformly from $\{1,\dots ,q\}$. Equivalently, every compatible spin configuration has conditional probability $q^{-k(\omega )}$, and all incompatible configurations have probability zero. These are the two conditional laws of the Edwards-Sokal coupling.

If $x,y$ belong to the same open component, their colors agree surely; otherwise the two independent component colors agree with probability $1/q$. ThereforeRearranging proves the spin-connectivity identity for the Potts model:

Phase transitions refer to infinite-volume limits, not a single finite graph. The identity shows that connectivity is exactly the excess spin agreement above its independent-color baseline. More directly, use a wired random-cluster boundary condition and fix the corresponding boundary spin to color $1$. Its component has color $1$, while every finite component has a uniform color. In the compatible infinite-volume coupling,Consequently normalized spontaneous magnetization isThe onset of a wired infinite cluster and the onset of spontaneous magnetization in the Potts model are the same transition under $p=1-e^{-\beta }$, giving ${\beta }_c(q)=-\log (1-p_c(q))$. This also transfers any jump or continuity of that order parameter between the two descriptions; it makes no assumption about behavior exactly at the critical point.

The ferromagnetic Potts model on a labelled ring has $Q$ possible site states, $J>0$ and ${\sigma }_{L+1}={\sigma }_1$. Its spin-chain transfer matrix is $T=(e^{\beta J}-1)I+P$, with $P$ the $Q$-dimensional all-ones matrix. The canonical partition function is $(e^{\beta J}+Q-1{)}^L+(Q-1)(e^{\beta J}-1{)}^L$. The convention counts two bonds when $L=2$.

The ferromagnetic Potts model on a finite graph assigns $q$ possible states to each graph vertex and rewards neighbouring equal states by the displayed weight, for $\beta \ge 0$. The interaction sum counts each unoriented edge once. The Edwards-Sokal coupling identifies its spin clusters with the random-cluster model at $p=1-e^{-\beta }$ and integer $q\ge 2$.

On a finite graph $G=(V,E)$, the random-cluster model weights an open-edge configuration $\omega \subseteq E$ byThe partition function $Z_{p,q}$ is the sum of these weights, and $k(\omega )$ counts the open connected components of a graph, including isolated graph vertices. The endpoint laws at $p=0$ and $p=1$ are obtained by limits. The factor $q$ is the cluster weight in the random-cluster model.^[ref]

At $q=1$, the component factor disappears and the law is independent bond percolation. For other $q$, $p$ is a model parameter rather than necessarily the marginal opening probability of each edge. For a single edge joining two vertices, with $p=1/2,q=2$, the closed and open weights are $2$ and $1$, so the edge is open with probability $1/3$.

The random-cluster single-edge conditional probability describes the local mechanism. The Edwards-Sokal coupling connects integer cluster weights with the Potts model, while random-cluster boundary conditions specify how boundary vertices are identified before component counting.

R. G. Edwards, identified as Robert G. Edwards in the publisher byline, coauthored the Edwards-Sokal coupling with A. D. Sokal. The construction combines spin and bond variables in one joint probability law, connecting the Potts model and the random-cluster model. This supplies the probabilistic basis for changing entire clusters in a Monte Carlo method.^[ref]

For integer $q\ge 2$, given a random-cluster configuration, its components receive independent uniform colors under the Edwards-Sokal coupling. Two vertices have equal Potts spins with probability $1$ if connected and $1/q$ otherwise. Averaging proves the identity. With a wired component fixed to color $1$, the same argument gives normalized spontaneous magnetization $(q{\pi }^1({\sigma }_0=1)-1)/(q-1)={\varphi }^{\mathrm{w}}(0\leftrightarrow \infty )$ in infinite volume. Thus its onset is the wired random-cluster critical probability, with dimensionless Potts model coupling $K_c=-\log (1-p_c)$ in the convention of the Edwards-Sokal coupling.

For $q=1$, the equal-spin probability is identically one and the first identity reduces to $1=1$. The normalized magnetization expression divides by $q-1$ and therefore does not apply to that degenerate one-color case. The conversion to a physical Ising model inverse temperature must also account for the interaction convention.