Table of Contents
Fetching ...

A characterisation of the continuum Gaussian free field in $d \geq 2$ dimensions

Juhan Aru, Ellen Powell

Abstract

We prove that under certain mild moment and continuity assumptions, the $d$-dimensional Gaussian free field is the only stochastic process in $d\geq 2$ that is translation invariant, exhibits a certain scaling, and satisfies the usual domain Markov property. Our proof is based on a decomposition of the underlying functional space in terms of radial processes and spherical harmonics.

A characterisation of the continuum Gaussian free field in $d \geq 2$ dimensions

Abstract

We prove that under certain mild moment and continuity assumptions, the -dimensional Gaussian free field is the only stochastic process in that is translation invariant, exhibits a certain scaling, and satisfies the usual domain Markov property. Our proof is based on a decomposition of the underlying functional space in terms of radial processes and spherical harmonics.

Paper Structure

This paper contains 12 sections, 10 theorems, 56 equations.

Key Result

Theorem \oldthetheorem

Let $d \geq 2$ and $\mathbb{B} \subseteq \mathbb{R}^d$ denote the open unit ball. Suppose that $h$ is a random Schwartz distribution with support on $\mathbb{B}$, i.e. a random continuous functional $(h,f)$ indexed by $f \in C_c^\infty(\mathbb{R}^d)$ and zero as soon as $f$ has support oustide of $\

Theorems & Definitions (26)

  • Definition 1: Gaussian free field
  • Definition \oldthetheorem: Domain Markov Property (DMP) with scaling for balls
  • Theorem \oldthetheorem: Characterisation of the $d$-dimensional GFF
  • Remark \oldthetheorem
  • Definition \oldthetheorem: Scaling function
  • Proposition \oldthetheorem
  • proof
  • Corollary \oldthetheorem
  • Lemma \oldthetheorem
  • proof : Proof of \ref{['4mom_av']}
  • ...and 16 more