🔢 mathematics

Sharp gradient integrability for $(s,p)$ -Poisson type equations

This paper establishes optimal local $W^{1,q}$ -regularity and quantitative gradient estimates for weak solutions to fractional $p$ -Laplacian type equations with $L^r$ right-hand sides, while providing a counterexample to confirm the optimality of the exponent $q$ .

Original authors: Verena Bögelein, Frank Duzaar, Naian Liao, Kristian Moring

Published 2026-02-10

📖 3 min read🧠 Deep dive

CC BY 4.0

Original authors: Verena Bögelein, Frank Duzaar, Naian Liao, Kristian Moring

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 you are a master chef trying to understand how heat spreads through a complex, multi-layered cake. You know that if you turn up the oven (the "input" or force), the temperature (the "solution") will change. But how exactly does the intensity of the heat relate to how smoothly the temperature changes across the cake?

This mathematical paper is essentially a high-level study of that relationship, but instead of heat in a cake, it’s about "energy" and "smoothness" in complex, non-local systems.

Here is the breakdown in everyday language:

1. The Core Problem: The "Smoothness" Mystery

In mathematics, we often deal with equations that describe how things spread out—like heat, or how a population moves. We have a "force" (the right-hand side of the equation, called $f$ ) and a "result" (the solution, $u$ ).

The big question is: If the force is somewhat steady and predictable, how smooth is the resulting movement?

If the force is a bit "jittery" (it belongs to a certain mathematical space), does the resulting movement stay smooth, or does it become jagged and wild? This paper provides a "sharp" answer, meaning they have found the exact mathematical limit where smoothness breaks down.

2. The "Fractional" Twist: The Telepathic Cake

Most classical math assumes things only affect their immediate neighbors (like heat moving from one molecule to the next). This is "local" math.

However, this paper studies "fractional" equations. In a fractional world, things are "non-local." Imagine if a molecule on the left side of the cake could instantly feel the temperature of a molecule on the far right side without the heat traveling through the middle. It’s like the molecules are telepathic. This makes the math incredibly difficult because you can't just look at what's happening "right here"; you have to account for the "Tail"—the influence of everything happening far away.

3. The "p-Laplacian": The Stubborn Material

The paper also deals with "p-type" equations. In standard math, materials react linearly (double the force, double the movement). But in "p-type" math, the material is stubborn or non-linear.

Think of it like trying to push a heavy object through different substances. Pushing through water is easy; pushing through thick honey is much harder and requires a different kind of effort. The "p" represents the "thickness" or the "stubbornness" of the medium.

4. The "Calderón-Zygmund" Goal: The Perfect Scale

The authors are performing what is called "Nonlinear Calderón-Zygmund theory."

Think of this as a Universal Scaling Law. If you know the "roughness" of the input, the Calderón-Zygmund theory gives you a perfect formula to predict the "roughness" of the output. The authors have successfully created this formula for these "telepathic, stubborn" systems. They didn't just find a general rule; they found the sharpest possible rule—the one that cannot be improved even by a tiny fraction.

Summary Metaphor: The Rugged Landscape

Imagine you are driving a car over a landscape.

The Force ( $f$ ) is the ruggedness of the terrain (the rocks and bumps).
The Solution ( $u$ ) is the path your car takes.
The Gradient ( $\nabla u$ ) is how much your steering wheel has to jerk around.

This paper proves that if the terrain has a certain level of "bumpiness," we can predict exactly how much your steering wheel will jerk. Even if the terrain is "telepathic" (the bumps far away affect you instantly) and "stubborn" (the car's tires react weirdly to the bumps), the math holds up. They have found the ultimate "speed limit" for how much that steering wheel can shake.

This paper, titled "Sharp Gradient Integrability for $(s, p)$ -Poisson Type Equations," provides a comprehensive regularity theory for the fractional $p$ -Laplacian operator with variable coefficients. It establishes a nonlinear version of the classical Calderón-Zygmund theory, specifically quantifying how the integrability of the right-hand side $f$ dictates the integrability of the gradient $\nabla u$ .

1. The Problem

The authors investigate the regularity of local weak solutions to the $(s, p)$ -Poisson type equation:
$(-\mathbf{a}\Delta_p)_s u := \text{p.v.} \int_{\mathbb{R}^n} \frac{a(x, y)|u(x) - u(y)|^{p-2}(u(x) - u(y))}{|x - y|^{n+sp}} \, dy = f(x) \text{ in } \Omega$
where $s \in (0, 1)$ is the fractional order, $p > 1$ is the power of the operator, and $a(x, y)$ is a measurable, symmetric coefficient satisfying ellipticity, boundedness, and a specific Höllder-type continuity condition.

The central challenge is that for fractional operators, the "gradient" is not inherently part of the weak solution definition (unlike the local $p$ -Laplacian). Furthermore, the non-local nature of the operator introduces "tail terms"—contributions from the behavior of the function $u$ far outside the local domain—which must be accounted for in all estimates.

2. Methodology

The paper employs a sophisticated multi-stage approach to bridge the gap between the fractional energy space $W^{s,p}$ and higher-order Sobolev spaces $W^{1,q}$ :

Finite Difference Scheme: Since fractional equations cannot be differentiated directly to obtain an equation for the gradient, the authors use second-order finite differences ( $\tau_h^2 u$ ) to probe differentiability.
Comparison Principle: A core technique involves comparing the solution $u$ to a "fractional $p$ -harmonic replacement" $v$ (the solution to the homogeneous problem with the same boundary data). This allows the authors to separate the effects of the inhomogeneity $f$ from the operator's structural properties.
Atomic Decomposition & Scaling: To handle the low integrability of $f$ , the authors implement a refined finite difference scheme at finer scales, effectively using an "atomic decomposition" to manage increments.
Stopping Time & Vitali Covering: To prove the sharp $L^q$ estimates, the authors use a stopping-time argument to identify "bad" sets where the gradient is large. They then apply a Vitali-type covering lemma to these super-level sets and use a Calderón-Zygmund-type decomposition to integrate the gradient over these sets.

3. Key Contributions and Results

The paper provides three major tiers of results:

A. Sharp Gradient Integrability (Theorem 1.1):
The primary result is the establishment of the optimal exponent $q$ such that if $f \in L^r_{\text{loc}}(\Omega)$ , then $\nabla u \in L^q_{\text{loc}}(\Omega)$ . The exponent is given by:
$q = \frac{rn(p-1)}{n - r(p-1)(sp'-1)}$
The authors prove this is sharp by providing an explicit counterexample showing that the integrability cannot be improved.

B. Higher Differentiability (Theorem 1.3 & 4.4):
The authors show that if $f$ has sufficient integrability, the solution $u$ actually possesses higher fractional differentiability, specifically $u \in W^{1+\gamma, p}_{\text{loc}}(\Omega)$ for some $\gamma > 0$ . This result holds even for non-differentiable variable coefficients $a(x, y)$ .

C. Quantitative Estimates with Nonlocal Tails:
Unlike local theory, all estimates (for the gradient and the higher-order derivatives) include a nonlocal tail term. This term quantifies how the values of $u$ at infinity affect the local regularity, a fundamental requirement for fractional calculus.

4. Significance

This work is significant for several reasons:

Completes the Analogy: It completes the analogy between the classical $p$ -Laplacian and the fractional $p$ -Laplacian in the context of Calderón-Zygmund theory.
Non-Divergence Form Handling: While much of the existing nonlinear Calderón-Zygmund theory focuses on divergence-form equations, this work provides a direct approach for non-divergence form inhomogeneities, which is useful in settings where standard tools (like the Bogovskiĭ operator) fail.
Mathematical Rigor in Nonlocal Settings: It provides the most refined known bounds for the integrability of the gradient in the fractional setting, accounting for the complex interplay between the fractional parameter $s$ , the power $p$ , and the dimension $n$ .