Gaussian free fields on Hamming graphs and
lattice spin systems · 2025
Read Gaussian free fields on Hamming graphs and by lattice spin systems (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-24199.txt.
Introduction
Shuhei Mano studies a discrete Gaussian free field on Hamming graphs. Such fields are random surfaces with Gaussian fluctuations. Hamming graphs arise from strings of symbols. Two vertices interact according to their Hamming distance alone. That sets this setting apart from the usual integer lattice. Nearest-neighbor interactions rule there instead.
Spectral and combinatorial tools drive the analysis. In high dimension or large alphabet limits, the field converges to independent Gaussians. A rescaled version behaves oppositely in the massless limit. It becomes perfectly correlated. One specific model gets close attention. Its interaction weights follow a binomial law in the distance. A double-scaling limit emerges when coupling ties to mass. Correlations there are convex in Hamming distance. At each distance, the correlation equals a simple reciprocal count.
Catalog record
| CREATOR | lattice spin systems |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-24199
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Gaussian free fields on Hamming graphs and lattice spin systems Abstract Abstract. We discuss a class of discrete Gaussian free fields on Hamming graphs, where interactions are determined solely by the Hamming distance between vertices. The purpose of examining this class is that it differs significantly from the commonly discussed spin system on the integer lattice with nearest-neighbour interactions. After introducing general results on the partition function and covariance for the class of Gaussian free fields, we present detailed properties of some specific models. Group-theoretic arguments and the Fourier transform give some explicit results. Key Words and Phrases. Fourier transform, Gibbs measure, Green function, group representation, random walk, statistical mechanics 2020 Mathematics Subject Classification Numbers. 60G60, 05E30, 60B15, 82B20 1 Introduction Consider a finite set \mathcal{X}=\{0,1,\ldots,n-1\}^{d} , where d,n\geq 2 . Set \rho(x,x^{\prime})=|\{i:x_{i}\neq x^{\prime}_{j}\}| for x=(x_{i})\in\mathcal{X} and x^{\prime}=(x^{\prime}_{i})\in\mathcal{X} , which is called the Hamming distance . The Hamming distance induces a relation \mathcal{X}\times\mathcal{X} by (x,x^{\prime})\in R_{i}~\Leftrightarrow~\rho(x,x^{\prime})=i\quad\text{for}~i\in\{0,1,\ldots,d\}. An undirected graph (\mathcal{X},R_{1}) with vertices \mathcal{X} and edges R_{1} is called the Hamming graph H(d,n) . The Hamming graph is regular. The number of equidistant vertices is denoted by \kappa_{i}:=|\{x^{\prime}\in\mathcal{X}:\rho(x,x^{\prime})=i\}|=\binom{d}{i}(n-1)^{i},\quad i\in\{0,1,\ldots,d\} (1) for any x\in\mathcal{X} , where \kappa_{1} is the degree of a vertex, namely, the number of adjacent vertices. The Hamming graph is a fundamental object in algebraic combinatorics, and group-theoretic approach is considered to be effective [ 2 ] . In this paper, we consider Gaussian free fields on the Hamming graph. The Hamming graph is quite different from the integer lattice \mathbb{Z}^{d} with edges between nearest-neighbour vertices. For example, the diameter of H(d,n) is d , while that of a cube with side length n is nd . Noting that the number of vertices is the same in both graphs, we can say that the Hamming graph is more densely connected than the integer lattice and “smaller” than the integer lattice. Let \partial\subset{X} be a distinguished set of vertices, called the boundary of the graph, and set \mathcal{Y}=\mathcal{X}\setminus\partial . In this paper, we discuss a class of discrete Gaussian free fields on H(d,n) , denoted by (g_{x})_{x\in\mathcal{X}} , defined by the Hamiltonian H_{\mathcal{Y}}((\xi_{x})_{x\in\mathcal{X}}):=\frac{1}{4}\sum_{i=1}^{d}J_{i}\sum_{(x,x^{\prime})\in R_{i}}(\xi_{x}-\xi_{x^{\prime}})^{2}+\frac{m^{2}}{2}\sum_{x\in\mathcal{Y}}\xi_{x}^{2},\quad\mathcal{Y}\subset\mathcal{X}, (2) where (\xi_{x})_{x\in\mathcal{X}}\in\mathbb{R}^{\mathcal{X}} are values of the Gaussian random variables called spins . The Hamiltonian tends to be small if spins take similar values. Here, we consider the interaction between spins \xi_{x} and \xi_{x^{\prime}} solely depends on the Hamming distance i=\rho(x,x^{\prime}) through weights J_{i}\geq 0 satisfying \sum_{i=1}^{d}\kappa_{i}J_{i}\leq 1 . The parameter m\geq 0 is called mass . The centred Gaussian free field satisfying the boundary condition \mathbf{E}[g_{x}]=0 , x\in\partial has the Gibbs measure with parameter \beta>0 : \mu_{\mathcal{Y}}(S)=\frac{1}{Z_{\mathcal{Y}}}\int_{S}e^{-\beta H_{\mathcal{Y}}((\xi_{x})_{x\in\mathcal{X}})}\prod_{x\in\mathcal{Y}}d\xi_{x}\quad\text{for~each}\quad S\in\mathcal{B}, (3) where \mathcal{B} is the Borell set over the hyperplane \mathcal{P}_{\mathcal{Y}}=\{(\xi_{x})_{x\in\mathcal{X}}\subset\mathbb{R}^{\mathcal{X}}:\xi_{x}=0~\text{for~all}~x\in\partial\}. (4) The normalisation constant Z_{\mathcal{Y}} is called the partition function. In studies of spin systems, \beta is called the inverse temperature. The effect of this scaling is trivial, since ( 2 ) is a homogeneous polynomial; more precisely, the scaling makes the variance of the Gaussian free field 1/\beta times. Research on discrete Gaussian free fields has a long history; however, much of the research has focused on their properties on the integer lattice \mathbb{Z}^{d} , particularly in two dimensions ( d=2 ). In particular, there has been considerable research in the context of Liouville quantum gravity and Schramm–Loewner evolution (see [ 3 ] for a book-length treatment), as well as in the context of extreme values [ 4 ] . On the other hand, there has been relatively little research on Gaussian free fields on graphs other than \mathbb{Z}^{d} . The few exceptions are, for example, Gaussian free fields on compact manifolds [ 5 ] and the discrete torus [ 1 ] . To the best of the author’s knowledge, there are no studies on Gaussian free fields on the Hamming graphs; however, expansions of Gaussian free fields on the hypercube, H(d,2) , by independent standard Gaussian random variables were extensively studied in [ 6 , 10 ] . In this paper, we discuss discrete Gaussian free fields on H(d,n) with homogeneous interactions determined solely by the Hamming distance. In Section 2 , we introduce the random walk representation for studying the Gaussian free fields. First, we introduce a random walk on a Hamming graph, where the transition probability is determined solely by the Hamming distance between vertices, and identify its Green function as the covariance of the Gaussian free fields described above. We then discuss the massive ( m>0 ) cases without boundary ( \partial=\emptyset ), and the massless ( m=0 ) cases with boundary. Here, the boundary refers to the outside of a set of vertices equidistant from the origin. For the former, we derive an expansion of the Green function in terms of an orthonormal system and provide an expression for the Gaussian free field in terms of the standard Gaussian distribution. For the latter, we provide an expression for the variance of the Gaussian random variable at the origin. In Section 3 , we discuss general properties of the Gaussian free fields using the random walk representation introduced in Section 2 and the derived results. In particular, we discuss the massive cases without boundary. We provide explicit expressions for the partition function and demonstrate the convergence of the free energy (defined in Section 3 ) and the limits for large d or n . We show that in the massive limit ( m\to\infty ), scaled Gaussian random variables converge weakly to the independent standard Gaussian distributions. We then discuss covariance of the Gaussian free fields. We show that in the massless limit ( m\to 0 ), Gaussian random variables diverge while remaining perfectly correlated. We show that in the limits of large d or n , the Gaussian random variables converge weakly to the independent standard Gaussian distributions. We also show that, for large n , the covariance decays exponentially with the Hamming distance, irrespective of interactions among spins. Section 4 provides detailed descriptions of three specific lattice spin systems that help clarify the properties of the Gaussian free fields derived in Section 3 . Section 4.1 considers a model in which the interaction weights between all vertices are equal. We provide explicit results for all cases: massive without boundary, massive with boundary, and massless with boundary. In Section 4.2 , we consider a model in which only adjacent vertices have equal-weight interactions, i.e., the nearest-neighbour interactions. We show that for the massless without boundary case, for large d or n , the covariance decays exponentially with the Hamming distance. Section 4.3 considers a model in which interactions exist between all pair of the vertices, but the weights vary monotonically with the distance between vertices. We show that convergence to the limit d\to\infty is fastest when the weights are constant, i.e., when the model reduces to the first model. 2 The random walk representation In this section, we introduce random walks which are useful to analyse Gaussian free fields on the Hamming graph H(d,n) . See [ 3 , Section 1.1] , [ 7 , Sections 8.4,5] , and [ 8 , Section 13.1] for the random walk representation. See [ 2 , Chapter 1,2] for details about the Hamming graph and related notions. Consider a discrete-time Markov chain \{X_{t}:t\geq