Minkowski content construction of the CLE gasket measure

✨

This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Measuring the Unmeasurable

Imagine you are looking at a piece of Swiss cheese, but instead of holes, it's made of an infinite number of tangled, wiggly loops. This is what mathematicians call a CLE Gasket (Conformal Loop Ensemble). It's a shape that exists in the "fractal" world—a world where things are infinitely detailed, and if you zoom in, you see more loops, and if you zoom in again, you see even more.

The problem is: How do you measure the "size" (or mass) of this shape?

In the normal world, if you want to know the area of a square, you just multiply length by width. But a CLE Gasket is too jagged and complex for that. It's like trying to measure the length of a coastline with a ruler; the smaller your ruler, the longer the coastline gets.

For a long time, mathematicians knew this shape had a specific "mass" or "measure," but they could only find it using a very indirect, complicated method (like guessing the weight of a cloud by measuring the humidity of the air around it).

This paper says: "We can do better. We can measure this shape directly, just like we measure a table, by using simple counting methods."

The Five Ways to Count the "Dust"

The authors show that you can estimate the mass of this fractal shape using five different "approximation schemes." Think of these as five different ways to count the dust on a shelf.

The Box Count (The Grid Method):
Imagine laying a grid of tiny squares over your shape. Count how many squares touch the shape. As the squares get smaller and smaller, this count (adjusted for the size of the squares) tells you the mass.
- Analogy: Like counting how many pixels light up on a screen to see how big a picture is.
The "Double-Box" Count:
Similar to the first one, but you check if the shape touches a slightly larger square around the original one. This is a specific variation used in previous studies.
The Minkowski Content (The "Paint" Method):
Imagine you have a paintbrush with a width of $\delta$ . You paint a layer of paint around the entire shape. The amount of paint you use (the area of the painted region) tells you the mass. As your brush gets thinner ( $\delta \to 0$ ), the calculation becomes exact.
- Analogy: Measuring the volume of a sponge by seeing how much water it absorbs when you dip it in.
The Geodesic Ball Count (The "Walking" Method):
This is a clever twist. Instead of measuring distance in a straight line (Euclidean), you measure distance by "walking" along the shape itself. You count how many "steps" (balls) of a certain size it takes to cover the shape if you are forced to stay on the loops.
- Analogy: Measuring the size of a winding mountain trail by counting how many 1-mile markers fit on the path, rather than measuring the straight-line distance from the bottom to the top.
The Resistance Ball Count (The "Electricity" Method):
This treats the shape like an electrical circuit. It counts how many "resistors" (balls) are needed to cover the shape based on how electricity would flow through it.
- Analogy: Figuring out the size of a city's road network by seeing how many toll booths you'd need to block traffic effectively.

The Main Discovery: The paper proves that all five of these methods give you the exact same answer. They all converge to the same "canonical" measure. It's like weighing a bag of apples on a digital scale, a spring scale, and a balance scale, and finding they all agree perfectly.

Why Does This Matter? (The "Percolation" Connection)

One of the most exciting parts of the paper is what happens when you apply this to CLE6 (a specific type of loop pattern).

The Old Way: In 2013, mathematicians Garban, Pete, and Schramm found a way to measure this shape by looking at critical percolation (a model of how water flows through a porous rock or how a virus spreads through a population). They counted the number of "vertices" (dots) in a giant cluster.
The New Connection: This paper proves that the "indirect" measure they found in 2013 is exactly the same as the "direct" measure found in this paper.
The Metaphor: It's like discovering that the "weight" of a cloud calculated by a satellite (indirect) is exactly the same as the weight calculated by catching rain in a bucket (direct). This confirms that our mathematical models of how things spread and cluster are consistent.

This is crucial because it helps scientists understand how random walks (like a drunk person stumbling around) behave on these fractal shapes, which has applications in physics and material science.

The "Moments" (The "No-Spikes" Guarantee)

The paper also proves something called "finiteness of moments."

The Problem: In some weird fractal shapes, the mass can be "spiky." Imagine a shape that is mostly empty, but has one tiny, incredibly dense point that holds all the weight. If you try to measure the "average" weight, that one spike ruins the calculation.
The Result: The authors prove that for these CLE gaskets, the mass is smoothly distributed. There are no crazy spikes. The mass is spread out evenly enough that you can calculate averages, variances, and higher-order statistics without the math breaking down.
Analogy: It's the difference between a pile of sand (smooth, predictable) and a pile of sand with a single, invisible, infinitely heavy diamond hidden inside (chaotic, unpredictable). This paper proves the CLE gasket is a pile of sand, not a pile of sand with a diamond.

Summary

In short, Jason Miller and Yizheng Yuan took a mysterious, infinitely complex shape made of loops and showed us how to measure it directly using simple, intuitive tools (counting boxes, painting, walking, and electricity). They proved that all these tools agree, they connected this new measurement to old, famous physics models, and they showed that the shape is "well-behaved" enough to be studied with standard statistical tools.

They turned a ghostly, abstract mathematical concept into something tangible and measurable.

1. Problem Statement and Context

The Object of Study:
The paper focuses on the Conformal Loop Ensemble (CLE $_\kappa$ ) for $\kappa \in (4, 8)$ . CLE $_\kappa$ is a random collection of loops in a simply connected domain that arises as the scaling limit of interfaces in various 2D statistical mechanics models (e.g., percolation for $\kappa=6$ , FK-Ising for $\kappa=16/3$ ).

For $\kappa \in (4, 8)$ , the loops are self-intersecting and intersect the domain boundary.
The CLE gasket ( $\Upsilon$ ) is defined as the set of points in the domain that are not inside any of the loops. It is a random fractal set with a specific Hausdorff dimension $d_{CLE} = 2 - \frac{(8-\kappa)(3\kappa-8)}{32\kappa}$ .

The Core Problem:
A canonical, conformally covariant measure $\mu(\cdot; \Upsilon)$ on the CLE gasket was previously constructed indirectly by Miller and Schoug (MS24) using Liouville Quantum Gravity (LQG) and growth-fragmentation processes. However, this construction was non-constructive in the sense that it did not provide a direct geometric approximation scheme (like Minkowski content) to recover the measure from the set itself.

Goal: To show that the MS24 measure can be realized as the limit of several natural, direct approximation schemes, specifically Minkowski content and related counting measures.
Secondary Goal: To establish that the CLE $_6$ gasket measure coincides with the measure constructed by Garban, Pete, and Schramm (GPS13) as a scaling limit of critical percolation cluster counts.
Regularity: To prove that the measure has finite moments of all orders (previously only the first moment was known) and satisfies sharp regularity bounds.

2. Methodology

The authors employ a combination of probabilistic techniques, conformal invariance, and spatial independence arguments.

A. Approximation Schemes:
The paper defines five specific approximation schemes for the measure $\mu_k(B; \Upsilon)$ on a set $B$ :

Box Count: Counting dyadic squares of side $2^{-k}$ intersecting the gasket, renormalized by $2^{-k d_{CLE}}$ .
GPS Variant: Counting squares where the doubled square intersects the gasket (related to percolation cluster counting).
Euclidean Minkowski Content: The volume of the $\delta$ -neighborhood of the gasket, renormalized by $\delta^{d_{CLE}-2}$ .
Geodesic Metric Ball Count: The minimal number of balls of radius $\delta^{\alpha_g}$ (in the intrinsic geodesic metric) needed to cover the gasket.
Resistance Metric Ball Count: The minimal number of balls of radius $\delta^{\alpha_r}$ (in the intrinsic resistance metric) needed to cover the gasket.

B. Key Technical Tools:

Spatial Independence (AMY25a): The authors heavily rely on the "independence across scales" property established in a companion paper (Amy, Miller, Yuan). This allows them to treat the CLE structure in disjoint annuli as conditionally independent given the boundary conditions (link patterns).
Multichordal CLE: They utilize the theory of multichordal CLE (CLE conditioned on specific boundary connections) to analyze the local behavior of the gasket and establish absolute continuity between different CLE configurations.
Moment Bounds: A significant portion of the work involves proving that the measure $\mu$ has finite moments of all orders ( $E[\mu(K)^n] < \infty$ ). This is achieved by deriving sharp upper and lower bounds for the measure of small balls using conformal covariance and the probability of the gasket intersecting a ball.
Regularity Estimates: They prove that the measure is "regular" in the sense that it does not concentrate too much or too little mass in small balls (Propositions 1.4 and 1.5).

C. Proof Strategy for Convergence:

Comparability: First, they show that the approximation schemes $\mu_k$ and the true measure $\mu$ are comparable up to deterministic constants (i.e., $c_1 \mu \le \liminf \mu_k \le \limsup \mu_k \le c_2 \mu$ ).
Self-Similarity and Subsequences: Using the self-similarity of the CLE, they show that along subsequences, the ratio $\mu_k / \mu$ converges in probability to a constant.
Almost Sure Convergence: By refining the estimates and using the independence across scales, they upgrade convergence in probability to almost sure convergence.
Constant Identification: They prove that the scaling constants in the approximation schemes converge to a single deterministic constant $c$ , independent of the specific realization of the gasket.

3. Key Contributions and Results

Theorem 1.2 (Main Result):
For any nested CLE $_\kappa$ ( $\kappa \in (4, 8)$ ) and its gasket $\Upsilon$ , the canonical measure $\mu(\cdot; \Upsilon)$ is the almost sure limit of the approximation schemes (i)–(v) defined above. Specifically:
$\lim_{k \to \infty} \mu_k(B; \Upsilon) = c \mu(B; \Upsilon)$
almost surely for any Borel set $B$ with $\mu(\partial B) = 0$ . The constant $c$ depends only on $\kappa$ and the specific approximation scheme.

Proposition 1.4 (Upper Bound/Regularity):
The measure satisfies sharp upper bounds. For any compact set $K$ , the probability that the measure of a ball $B(z, r)$ deviates significantly from $r^{d_{CLE}}$ decays faster than any polynomial in $r$ . This implies the measure is "well-behaved" and does not have heavy tails.

Proposition 1.5 (Lower Bound/Regularity):
A stronger lower bound is established, showing that the measure of a specific region connecting two points on the boundary of a metric ball is bounded below by $r^{d_{CLE}+a}$ with high probability. This is crucial for proving the non-degeneracy of the measure.

Corollary: Identification with GPS Measure:
For $\kappa=6$ , the paper proves that the MS24 measure agrees with the measure constructed by Garban-Pete-Schramm (GPS13) as the scaling limit of critical percolation cluster counts. This resolves a long-standing question about the consistency of different constructions of the CLE $_6$ measure.

Corollary: Finite Moments:
The paper proves that for any compact set $K$ , the random variable $\mu(K)$ has finite moments of all orders ( $E[\mu(K)^n] < \infty$ for all $n$ ). Previously, this was only known for $n=1$ .

4. Significance and Impact

Direct Construction of the Measure: The paper moves the field from an indirect, abstract construction (via LQG) to a concrete, geometric construction (via Minkowski content). This makes the measure more accessible for analysis and simulation.
Unification of Models: By identifying the CLE $_6$ measure with the percolation cluster limit, the paper bridges the gap between the probabilistic construction of CLE and the discrete statistical mechanics origins of these models.
Foundations for Random Walks: The identification of the measure is a critical step for understanding the scaling limit of random walks on CLE gaskets. The authors note that this result is essential for proving that random walks on percolation clusters converge to canonical Brownian motion on the CLE $_6$ gasket (a result in a companion paper, ÐMMY26).
Regularity Theory: The proof of finite moments of all orders and the sharp regularity bounds (Propositions 1.4 and 1.5) are of independent interest. These bounds are necessary for analyzing the fine-scale geometry of the gasket and for defining other associated objects, such as the resistance metric and geodesic metric mentioned in the paper.
Methodological Advance: The techniques developed, particularly the use of "independence across scales" combined with multichordal CLE conditioning, provide a robust framework for analyzing other fractal measures arising in conformal probability.

In summary, this paper provides the rigorous geometric foundation for the canonical measure on the CLE gasket, confirming that it behaves exactly as one would expect from a "natural" volume measure on a fractal set, and linking it directly to discrete models and intrinsic metrics.