Gaussian Free Field¶
The Gaussian free field is a Gaussian random field whose covariance is the Green operator of a specified Laplacian.
Core Idea¶
The Gaussian free field is a random field whose observations are Gaussian and whose correlations come from a Laplacian's Green function. A boundary or pinning rule is needed to select the field.
Scope of Application¶
Probability and mathematical physics use graph and continuum versions to model fluctuations. Continuum observations are generally pairings with test functions, not exact point heights.
Boundary: An arbitrary Gaussian process has no required Green covariance; a continuum GFF is not normally an ordinary pointwise surface.
Clarity¶
State the domain, Laplacian, boundary rule, and whether observations are vertices or test-function pairings.
Manages Complexity¶
Mean and Green covariance determine the Gaussian law.
Abstract Reasoning¶
Fix the operator and boundary condition, invert it as a Green operator, then define centered Gaussian observations with that covariance. Without a pin or zero-mean rule, a common height shift can leave the covariance of absolute heights undefined.
Knowledge Transfer¶
Graph and continuum fields share the covariance design, but pointwise graph values do not automatically survive the continuum limit. Both are strict kinds of Stochastic Process: vertex heights or continuum test-function pairings form the indexed random family; pointwise continuum heights are not assumed.
For example, on a four-vertex path with the two endpoints pinned to zero, the two interior heights have covariance matrix \(\frac13\begin{pmatrix}2&1\\1&2\end{pmatrix}\), the inverse Dirichlet graph Laplacian.
Relationships to Other Abstractions¶
Current abstraction Gaussian Free Field Domain-specific
Parents (1) — more general patterns this builds on
-
Gaussian Free Field is a kind of Stochastic Process Prime
A GFF is a jointly Gaussian indexed random family, over vertices on a finite graph or test-function pairings in the continuum.
Hierarchy path (1) — routes to 1 parentless root
- Gaussian Free Field → Stochastic Process
Neighborhood in Abstraction Space¶
Gaussian Free Field sits in a sparse region of the domain-specific corpus (92nd percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Unclustered & Miscellaneous (2551 abstractions)
Nearest neighbors
- Closed Linear Operator — 0.81
- Dirichlet Eigenvalue — 0.80
- Limiting Absorption Principle — 0.80
- Heat Kernel Signature — 0.78
- Exponentially Modified Gaussian Distribution — 0.78
Computed from structural-signature embeddings · 2026-10-08