Besov regularity of solutions to the Dirichlet problem for the Bessel $(p,s)$-Laplacian

This paper establishes global Besov regularity estimates for weak solutions to the Dirichlet problem of a fractional $p$-Laplacian defined via the Riesz fractional gradient by combining Lions-Calderón spaces, Besov embeddings, and an adapted Nirenberg difference quotient method, yielding specific regularity indices that depend on the interplay between the order $s$ and the exponent $p$ in both superquadratic and subquadratic regimes.

The Gist

Imagine you are trying to smooth out a crumpled piece of paper, but with a twist: the paper isn't just connected to its immediate neighbors; it's connected to the whole room. If you pull one corner, the entire sheet ripples instantly, no matter how far away the other parts are. This is the world of non-local mathematics, where things don't just react to what's touching them, but to what's happening everywhere else.

This paper by Borthagaray, Del Pezzo, and Rueda Niño is about understanding how "smooth" or "rough" the solution to a specific type of mathematical puzzle is when you deal with these long-range connections.

Here is the breakdown of their work using everyday analogies:

1. The Problem: The "Ghostly" Rubber Sheet

The authors are studying a specific equation called the Bessel (p, s)-Laplacian. Let's break that down:

• The "Laplacian": Think of this as a rule for how a rubber sheet settles into its most comfortable, low-energy shape. Usually, if you push a spot on a sheet, the neighbors push back.
• The "Fractional" part (s): In this paper, the sheet is "ghostly." If you push a spot, it doesn't just talk to the neighbors; it talks to the whole sheet at once. The strength of this "ghostly" connection depends on a number called $s$ (between 0 and 1).
• The "p" part: This controls how "stiff" the sheet is.
  • If $p$ is large (superquadratic), the sheet is very stiff and resists bending sharply.
  • If $p$ is small (subquadratic), the sheet is more flexible and floppy.
• The "Riesz Gradient": This is the special tool the authors use to measure the "slope" of this ghostly sheet. It's different from the standard tools mathematicians usually use, allowing them to see the sheet's shape in a new, more precise way.

2. The Goal: How Smooth is the Solution?

When you solve this equation (find the shape of the sheet), you get a "solution." The big question is: How smooth is this solution?

• Is it a perfectly polished marble surface?
• Is it a jagged, rocky terrain?
• Is it somewhere in between, like sandpaper?

In math, we measure this "smoothness" using something called Besov Regularity. Think of Besov spaces as a ruler that doesn't just measure height, but measures how "jagged" or "fractal" the surface is at different scales.

3. The Method: The "Difference Quotient" Trick

To figure out the smoothness, the authors used a clever technique adapted from a mathematician named Savaré.

• The Analogy: Imagine you have a blurry photo of a landscape. To see if the mountains are sharp or fuzzy, you take two photos: one of the original landscape and one where you shifted the camera slightly to the right. Then, you subtract the two images.
• The Math: They did this mathematically. They took their solution, shifted it slightly (a "translation"), and looked at the difference.
• The Twist: Because the domain (the shape of the room) has walls (boundaries), you can't just shift the whole picture. You have to be careful near the edges. The authors developed a "localized translation" method—like shifting only a small patch of the photo while keeping the edges fixed—to see how the solution behaves right up against the wall.

4. The Big Discovery: It Depends on the "Stiffness"

The paper's main result is a set of rules that tell you exactly how smooth the solution will be, depending on the "stiffness" ($p$) and the "ghostly reach" ($s$).

• Scenario A: The Stiff Sheet ($p \ge 2$)
  If the sheet is stiff, the solution is quite smooth. The authors proved that the solution has a specific level of smoothness that gets better as the "ghostly reach" ($s$) increases. It's like a stiff rubber sheet that snaps back into a very orderly shape.

  • The Result: They found the solution belongs to a specific "smoothness class" (a Besov space), giving a precise formula for how smooth it is.

• Scenario B: The Floppy Sheet ($1 < p < 2$)
  If the sheet is floppy, it's harder to predict. The solution is still smooth, but the rules change. The smoothness depends on a different balance between the stiffness and the reach.

  • The Result: They found a different formula for the smoothness in this case. Interestingly, for very floppy sheets, the smoothness is limited by a "half-step" rule, meaning it's slightly less smooth than the stiff case in certain conditions.

5. Why Does This Matter?

You might ask, "Who cares about smooth ghostly rubber sheets?"

• Real World: These equations model things like fractures in materials (how a crack spreads through concrete), image processing (removing noise from a photo without blurring the edges), and even neural networks (how information flows through a brain-like structure).
• Computer Simulations: If you want a computer to simulate these phenomena, you need to know how smooth the answer is. If the answer is very smooth, you can use a coarse grid (fewer pixels) and get a good result. If it's jagged, you need a super-fine grid (millions of pixels), which takes forever to compute.
• The Payoff: By proving exactly how smooth the solution is, the authors give computer scientists a "map." They can now say, "We know the solution is this smooth, so we can build a faster, more efficient algorithm to solve it."

Summary

In short, these mathematicians took a complex, "ghostly" equation that describes how things interact over long distances. They used a clever "shift-and-compare" trick to prove exactly how smooth the answers are, depending on how stiff or flexible the system is. This knowledge is the key to building better, faster computer models for everything from cracking bridges to cleaning up blurry photos.

Technical Summary

Here is a detailed technical summary of the paper "Besov Regularity of Solutions to the Dirichlet Problem for the Bessel (p, s)-Laplacian" by Borthagaray, Del Pezzo, and Rueda Niño.

1. Problem Statement

The paper investigates the regularity of weak solutions to the Dirichlet problem for a specific class of non-local, quasi-linear elliptic operators defined via the Riesz fractional gradient. The problem is formulated in a bounded Lipschitz domain $\Omega \subset \mathbb{R}^N$ ($N \ge 1$) as:

$$\begin{cases} -\Delta^s_p u = f & \text{in } \Omega \\ u = 0 & \text{in } \Omega^c \end{cases}$$

where:

• $p \in (1, \infty)$ is the integrability exponent.
• $s \in (0, 1)$ is the fractional order.
• The operator is defined as $-\Delta^s_p u := -\text{div}_s (|\nabla^s u|^{p-2} \nabla^s u)$.
• $\nabla^s$ and $\text{div}_s$ denote the Riesz fractional gradient and divergence, respectively, introduced by Shieh and Spector (2015). This operator differs fundamentally from the standard fractional $p$-Laplacian (often defined via integral differences) because it relies on a distributional definition of the fractional gradient.
• $f$ is a source term belonging to a suitable dual space, typically $X^{-s, p'}(\Omega)$.

The primary goal is to establish global Besov regularity estimates for the weak solution $u$, specifically determining the smoothness index in the Besov scale $\dot{B}^\sigma_{p, \infty}(\Omega)$ based on the parameters $s$ and $p$.

2. Methodology

The authors employ a sophisticated analytical framework combining several advanced techniques:

• Functional Setting (Lions-Calderón Spaces): The analysis is grounded in the Lions-Calderón spaces $X^{s,p}(\Omega)$, which are shown to coincide with Bessel potential spaces $\Lambda^{s,p}(\Omega)$. These spaces are defined via the Riesz fractional gradient and are equivalent to complex interpolation spaces between $L^p$ and $W^{1,p}$.
• Difference Quotient Method (Savaré's Adaptation): The core of the proof adapts Nirenberg's difference quotient method, originally developed for local elliptic equations, to the fractional setting. This adaptation, pioneered by Savaré for Lipschitz domains, involves:
  • Localized Translations: Instead of global translations, the authors use localized translation operators $T_h$ weighted by a smooth cut-off function. This is crucial for handling the boundary behavior in Lipschitz domains.
  • Commutator Estimates: A key technical step involves analyzing the fractional gradient of the localized translation, $\nabla^s(T_h v)$. The authors derive an identity separating the localized translation of the gradient from a commutator term $C_{\phi, v}$. They prove that this commutator satisfies specific $L^p$ bounds dependent on the translation step $|h|$.
• Regularity of Functionals: The authors define a notion of $(T, D, \sigma)$-regularity for the energy functional $J(v) = \frac{1}{p}\int |\nabla^s v|^p - \langle f, v \rangle$. They demonstrate that the functional is regular with respect to the localized translations, which implies regularity for its minimizers (the weak solutions).
• Bootstrap Argument: The proof utilizes a recursive bootstrap argument. By establishing an initial regularity estimate, the authors iteratively improve the differentiability index of the solution, leveraging the monotonicity of the operator and the specific structure of the difference quotients.

3. Key Contributions

• Extension to Riesz Fractional Gradient: While previous works (e.g., Schikorra, Shieh, Spector) established local $C^{s+\alpha}$ regularity for the homogeneous case, this paper provides global Besov regularity estimates for the inhomogeneous Dirichlet problem with general $L^p$-type data.
• **Unified Framework for $p \ge 2$ and $1 < p < 2$:** The paper distinguishes between the superquadratic ($p \ge 2$) and subquadratic ($1 < p < 2$) regimes, deriving distinct regularity exponents for each.
• Variable Diffusivity Extension: The methodology is shown to extend to operators with variable diffusivity coefficients $A(x)$, provided $A$ is Lipschitz continuous.
• Interpolation Results: The authors derive intermediate regularity results using real interpolation theory (Tartar's nonlinear interpolation), providing mapping properties for the solution operator between fractional Sobolev and Besov spaces.

4. Main Results

The authors establish global Besov regularity $u \in \dot{B}^\sigma_{p, \infty}(\Omega)$ with quantitative bounds depending on the source data $f$. The regularity exponent $\sigma$ depends on the interplay between $s$ and $p$:

Case A: Superquadratic Regime ($p \ge 2$)

• High Differentiability ($s \in [1/p', 1)$):
  If $f \in B^{-s+1/p'}_{p', 1}(\Omega)$, then $u \in \dot{B}^{s + 1/p}_{p, \infty}(\Omega)$.
• Low Differentiability ($s \in (0, 1/p')$):
  If $f \in B^0_{p', 1}(\Omega)$, then $u \in \dot{B}^{s + \frac{s}{p-1}}_{p, \infty}(\Omega)$.

Case B: Subquadratic Regime ($1 < p < 2$)

• High Differentiability ($s \in [1/2, 1)$):
  If $f \in B^{-s+1/2}_{p', 1}(\Omega)$, then $u \in \dot{B}^{s + 1/2}_{p, \infty}(\Omega)$.
• Low Differentiability ($s \in (0, 1/2)$):
  If $f \in B^0_{p', 1}(\Omega)$, then $u \in \dot{B}^{2s}_{p, \infty}(\Omega)$.

Summary Formula:
The results can be summarized as:
$$u \in \dot{B}^{s + \min\{\frac{s}{p-1}, \frac{1}{p}\}}_{p, \infty}(\Omega) \quad \text{for } p \ge 2$$
$$u \in \dot{B}^{s + \min\{s, \frac{1}{2}\}}_{p, \infty}(\Omega) \quad \text{for } 1 < p < 2$$
(Note: The specific exponents in the low $s$ regimes reflect the critical thresholds where the regularity behavior changes.)

5. Significance and Impact

• Theoretical Advancement: This work bridges the gap between the theory of local $p$-Laplacians and non-local fractional operators defined via Riesz gradients. It confirms that the "Bessel (p, s)-Laplacian" possesses regularity properties analogous to classical elliptic equations, despite the non-local nature of the operator.
• Numerical Analysis: The derived Besov regularity estimates are critical for the convergence analysis of numerical schemes. The authors note in the introduction that these results are a prerequisite for a forthcoming Finite Element Method (FEM) analysis of problem (1.1), as convergence rates in FEM are directly tied to the smoothness of the exact solution.
• Generality: By handling Lipschitz domains (rather than requiring smooth boundaries) and variable diffusivity, the results are applicable to a broader range of physical and engineering problems, including fracture mechanics, image processing, and non-local elasticity, where domain boundaries are often non-smooth.
• Methodological Innovation: The successful adaptation of Savaré's localized translation technique to the Riesz fractional gradient setting provides a robust template for analyzing other non-local quasi-linear operators.

In conclusion, the paper provides a rigorous and comprehensive regularity theory for a significant class of non-local operators, establishing precise global smoothness estimates that are essential for both theoretical understanding and practical numerical approximation.