🔢 mathematics

On Weiner criterion for massiveness on weighted graphs

This paper investigates $p$ -harmonic functions on infinite weighted graphs by characterizing $p$ -massiveness through the non-uniqueness of bounded Dirichlet solutions and establishing a Wiener-type criterion at infinity for such sets under volume doubling and weak $(1,p)$ -Poincaré inequality conditions.

Original authors: Lu Hao

Published 2026-03-31

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Lu Hao

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ 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

Imagine a vast, infinite city made of dots (vertices) connected by roads (edges). Some roads are wider or more important than others, represented by "weights." In this city, there are rules about how "energy" flows from one dot to another. Mathematicians call these rules $p$ -harmonic functions. Think of these functions as the temperature distribution in the city: if the temperature at a dot is the average of its neighbors, it's in a state of perfect balance (harmonic).

This paper, written by Lu Hao, asks a very specific question about this infinite city: Can we find a "safe zone" (a massive set) where a temperature can exist that is strictly between 0 and 1, without ever touching the boundaries?

Here is a breakdown of the paper's ideas using simple analogies:

1. The Core Concept: The "Massive" Set

Imagine you have a giant, infinite garden (the graph). You want to know if there is a specific patch of flowers (a subset of the garden) that is "massive."

What does "massive" mean here? It doesn't mean heavy. It means "big enough to matter."
The Test: Can you create a temperature map where:
- The temperature inside the patch is always between 0 (frozen) and 1 (boiling).
- The temperature is perfectly balanced everywhere inside the patch (no heat is being created or destroyed).
- But, right outside the patch (the boundary), the temperature is stuck at 1 (boiling).
The Result: If you can do this, the patch is massive. If you can't, the patch is "too small" or "leaky" to sustain its own unique temperature against the boiling outside.

2. The Big Discovery: Uniqueness vs. Chaos

The paper proves a fascinating link between this "massive" concept and a classic math problem called the Dirichlet Problem.

The Problem: You set the temperature on the boundary of a region. Can you predict exactly what the temperature will be inside?
The Finding:
- If the region is NOT massive, the answer is simple: There is only one correct way to fill in the temperatures inside. The system is predictable.
- If the region IS massive, the system breaks down. There are multiple valid ways to fill in the temperatures. The system has "chaos" or "freedom."
Analogy: Imagine a room with a heater. If the room is small (not massive), the temperature settles into one specific pattern. If the room is a massive, infinite cave, you could theoretically have different stable temperature patterns coexisting, depending on how the heat was introduced in the distant past.

3. The "Wiener Criterion": The Infinite Ruler

How do you tell if a shape is "massive" without trying to solve the temperature problem? You need a ruler. In the 1920s, mathematician Norbert Wiener created a rule for 3D space to check if a shape is "thick" enough.

Lu Hao extends this rule to these infinite weighted graphs. He creates a Wiener Criterion, which is essentially a sum of "capacities."

The Analogy: Imagine you are standing at the center of a target. You look at the "holes" in the target (the parts of the graph outside your region) at increasing distances.
The Rule: You calculate a "resistance score" for these holes at every distance step.
- If the sum of these scores diverges (goes to infinity), the holes are too "thick" or "numerous." The region is not massive. The heat from the boundary will always leak in and dominate.
- If the sum converges (stays finite), the holes are "thin" or "sparse" enough. The region is massive. It can hold its own unique temperature.

4. The Conditions: The City Must Be "Well-Behaved"

To make this ruler work, the city (the graph) can't be just any random mess. It needs to follow two rules:

Volume Doubling (VD): If you double the size of a neighborhood, the number of dots doesn't explode uncontrollably; it just grows by a predictable factor. (Like how a sphere's volume grows predictably in 3D space).
Poincaré Inequality: This ensures that if you walk through the city, you can't have sudden, impossible jumps in temperature without paying an "energy cost." It keeps the city smooth and connected.

5. The "Energy" Twist ( $D_p$ -Massiveness)

The paper also distinguishes between "massive" and " $D_p$ -massive."

Massive: Just needs some solution.
$D_p$ -Massive: Needs a solution that doesn't use up infinite "energy."
The Analogy: Being "massive" is like having a lightbulb that stays on. Being " $D_p$ -massive" is like having a lightbulb that stays on without draining the entire power grid of the universe. The paper proves that for a region to be " $D_p$ -massive," it must contain a smaller, "non-parabolic" (non-leaky) core.

6. Real-World Examples

The author tests these theories on grids (like $\mathbb{Z}^d$ , which is like a 3D checkerboard extended infinitely).

The "Thorn" Example: Imagine a shape that looks like a long, thin needle sticking out into infinity.
- If the needle is too thick (grows fast), it blocks the "leakage," and the space around it is massive.
- If the needle is too thin, the space around it is not massive.
The "Line" Example: In a 3D grid, if you remove a single line (1D), the rest of the space is massive. But if you remove a plane (2D), it might not be. The dimension matters!

Summary

This paper is a bridge between geometry (how big and shaped the holes are) and physics (how heat/energy flows). It gives us a precise mathematical formula (the Wiener Criterion) to look at an infinite, complex network and say: "Yes, this specific part of the network is big and solid enough to sustain its own unique identity, or no, it's too small and will be swallowed by the outside world."

It's like having a magic magnifying glass that tells you if a shadow is deep enough to hide in, or if it's just a faint smudge that the light will wash away.

1. Problem Statement

The paper investigates the theory of $p$ -harmonic functions on infinite, connected, and locally finite weighted graphs. The central problem is to characterize the $p$ -massiveness of a subset $\Omega \subset V$ .

$p$ -Massiveness: A set $\Omega$ is defined as $p$ -massive if there exists a non-constant bounded $p$ -harmonic function $u$ on $\Omega$ such that $0 < u < 1$ in $\Omega$ and $u=1$ on the complement $\Omega^c$ .
$D_p$ -Massiveness: A stronger condition where the function $u$ also has finite $p$ -Dirichlet energy ( $D_p(u) < \infty$ ).
Core Question: Under what geometric and analytic conditions on the graph and the set $\Omega$ is $\Omega$ $p$ -massive? Specifically, the paper aims to establish a Wiener-type criterion at infinity to determine massiveness, analogous to classical results in Euclidean space ( $\mathbb{R}^d$ ) and Riemannian manifolds.

2. Methodology

The author employs a combination of nonlinear potential theory, geometric analysis, and probabilistic intuition adapted to the discrete setting.

Analytic Framework: The study utilizes the $p$ -Laplacian operator $\Delta_p$ on weighted graphs. The analysis relies heavily on the $p$ -Dirichlet energy and $p$ -capacity ( $cap_p$ ), which measure the "size" of sets in terms of the energy required to maintain a potential difference.
Geometric Assumptions: To derive the Wiener criterion, the paper assumes the weighted graph $(V, \mu)$ $(V, μ)$ satisfies:
1. Volume Doubling (VD): $\mu(B(x, 2r)) \leq C_D \mu(B(x, r))$ .
2. Weak $(1, p)$ -Poincaré Inequality ( $P_p$ ): Controls the oscillation of functions by their gradients.
3. Condition $(p_0)$ : A lower bound on edge weights relative to vertex weights, ensuring local connectivity properties.
Key Techniques:
- Comparison Principles: Used to compare sub/super-harmonic functions to establish uniqueness and existence.
- Green's Formula & Energy Identities: Used to relate capacity to flux across level sets (Lemma 3.3).
- Iterative Schemes: The proof of the Wiener criterion involves constructing sequences of solutions on expanding domains (exhaustion of the graph) and analyzing their limits.
- De Giorgi-Nash-Moser Type Iteration: The necessity proof borrows techniques from Kilpeläinen and Malý (originally for $\mathbb{R}^n$ ), utilizing Sobolev inequalities and Harnack inequalities to estimate the decay of potentials near infinity.

3. Key Contributions and Results

A. Characterization of Massiveness and Liouville Properties

The paper establishes the equivalence between the geometric property of massiveness and the analytic property of non-uniqueness in boundary value problems.

Theorem 1.2: A set $\Omega$ is $p$ -massive (resp. $D_p$ -massive) if and only if the Dirichlet problem on $\Omega$ with bounded boundary data admits more than one bounded (resp. finite energy) solution.
Theorem 1.3 (Liouville Property): The graph admits a non-constant bounded $p$ -harmonic function if and only if there exist two disjoint $p$ -massive subsets. This extends previous results by Holopainen and Soardi from unweighted graphs to general weighted graphs.

B. The Wiener Criterion at Infinity

This is the paper's primary contribution. It provides a necessary and sufficient condition for a set to be $p$ -massive based on the capacity of the set's intersection with annuli.

Theorem 1.7: Let $(V, \mu)$ $(V, μ)$ satisfy (VD) and ( $P_p$ $P_{p}$ ). For a connected set $\Omega$ $Ω$ , if $\Omega$ $Ω$ is $p$ $p$ -massive, then for some $x_0 \in \Omega$ $x_{0} \in Ω$ :
$\sum_{n=1}^{\infty} \left( \frac{cap_p(A_n, B_{n+1})}{cap_p(B_n, B_{n+1})} \right)^{\frac{1}{p-1}} < \infty$
where $B_n = B(x_0, 2^n)$ $B_{n} = B (x_{0}, 2^{n})$ and $A_n = \Omega^c \cap B_n$ $A_{n} = Ω^{c} \cap B_{n}$ .
- Converse: If the graph also satisfies condition $(p_0)$ , the convergence of this series implies $\Omega$ is $p$ -massive.
Significance: This generalizes the classical Wiener criterion (Itô & McKean for $p=2$ on $\mathbb{Z}^d$ ) to arbitrary $p > 1$ and general weighted graphs. It links the "thickness" of the complement at infinity to the solvability of the Dirichlet problem.

C. Criterion for $D_p$ -Massiveness

Theorem 1.9: A subset $\Omega$ is $D_p$ -massive if and only if there exists a non-parabolic subset $\Omega_1 \subset \Omega$ such that $cap_p(\Omega_1, \Omega) < \infty$ . This connects finite energy solutions to the concept of non-parabolicity (recurrence vs. transience of random walks).

D. Applications to $\mathbb{Z}^d$

The paper applies the general theory to the integer lattice $\mathbb{Z}^d$ with standard weights:

Thorn Sets: For a "thorn" set $T$ defined by a growth function $f(n)$ , the paper derives the specific condition for $p$ -massiveness:
$\sum_{n=1}^{\infty} \left( \frac{f(2^n)}{2^n} \right)^{\frac{d-p-1}{p-1}} = \infty$
Coordinate Axes: It analyzes the massiveness of the complement of a 1D axis in $\mathbb{Z}^d$ , showing that for $d = p+1$ , the complement is not $p$ -massive, while for $d > p+1$ , it is $p$ -massive but not $D_p$ -massive (highlighting the distinction between bounded solutions and finite energy solutions).

4. Significance

Unification of Discrete and Continuous Theory: The paper successfully bridges the gap between potential theory on Riemannian manifolds (where Wiener criteria are well-established) and discrete weighted graphs. It proves that the geometric-analytic structure (VD + Poincaré) is sufficient to carry over deep results like the Wiener criterion to the discrete domain.
Generalization of $p$ -Harmonic Theory: By handling general $p > 1$ and weighted graphs, the results extend the classical $p=2$ (linear) theory and the unweighted graph results of Holopainen and Soardi.
Clarification of Massiveness Types: The paper rigorously distinguishes between $p$ -massiveness (bounded solutions) and $D_p$ -massiveness (finite energy solutions), providing specific criteria for each. This is crucial for understanding the behavior of random walks and diffusion processes on graphs.
Methodological Advancement: The adaptation of the Kilpeläinen-Malý iteration technique to the discrete setting demonstrates the robustness of nonlinear potential theory methods, offering a template for future research on discrete geometric analysis.

In summary, Lu Hao's paper provides a comprehensive framework for determining the "size" of infinite sets on weighted graphs through the lens of $p$ -harmonic functions, establishing a discrete Wiener criterion that is both theoretically profound and applicable to specific lattice structures.