🔢 mathematics

Towards gradient Hölder regularity for singular fractional $p$ -Laplace equations

This paper establishes the local gradient Hölder regularity for weak solutions of singular fractional $p$ -Laplace equations in the range $1 < p < 2$ and $0 < s < 1$ , providing a partial resolution to the open $C^{1,\alpha}$ problem in the singular regime.

Original authors: Chao Zhang

Published 2026-08-18

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Chao Zhang

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

In the vast landscape of mathematics, there is a branch dedicated to understanding how things change smoothly. Imagine a landscape of hills and valleys, where the height at any point represents a value like temperature or pressure. Mathematicians study equations that describe how these values settle into a stable shape. Often, these shapes are not jagged or broken; they are smooth, with slopes that change gradually rather than abruptly. This smoothness is crucial because it tells us that the physical world described by these equations behaves predictably. However, when the rules governing the change become extreme—specifically, when the force driving the change becomes very sensitive to tiny variations near zero—the mathematics becomes incredibly difficult. For decades, researchers have struggled to prove that solutions to these specific, difficult equations remain smooth, even when the underlying rules are harsh and singular.

This paper tackles one of those stubborn difficulties. It focuses on a class of equations known as the fractional p-Laplace equations, which model phenomena where the influence of a point extends far beyond its immediate neighbors, reaching out to distant parts of the system. The specific challenge addressed here arises when the parameter governing the system's sensitivity falls into a "singular" range, meaning the rules become infinitely sharp near zero. In this regime, standard mathematical tools often fail because the equations lose a property called uniform ellipticity, which usually guarantees smoothness. The author, Chao Zhang, proves that despite these harsh conditions, the solutions to these equations are indeed smooth enough to have a well-defined, continuously changing slope, provided the "fractional" nature of the equation is close enough to the standard, local version.

The core of the work is a proof that establishes a new level of regularity for these solutions. The author demonstrates that if the fractional order of the equation is sufficiently close to one, the solutions possess a gradient that is not just continuous, but Hölder continuous. In plain terms, this means the slope of the solution does not just exist; it changes in a controlled, predictable manner, without sudden jumps or wild oscillations. This result is significant because it resolves a major open question in the field for this specific singular range. Previous work had managed to prove similar smoothness for other ranges of parameters, but the singular case, where the equations are most delicate, had remained elusive. The paper shows that by carefully analyzing how the solution behaves when zoomed in, one can prove that it settles into a smooth pattern, effectively bridging the gap between the complex nonlocal behavior and the familiar smoothness of local equations.

To achieve this, the researcher developed a sophisticated method that involves looking at the equation through a series of magnifying glasses. The process begins by assuming the solution is not smooth and then zooming in repeatedly to see what happens. If the solution were truly rough, this zooming process would eventually reveal a pattern that contradicts known mathematical truths. The author introduces a clever way to normalize the equation during this zooming process, stripping away the noise and focusing on the essential shape. This normalization allows the researcher to compare the complex, nonlocal equation to simpler, local equations that are already well understood. By showing that the complex equation behaves like these simpler ones when viewed closely, the proof establishes that the solution must be smooth.

A critical part of the argument involves managing the "tails" of the equation. Because the fractional p-Laplace equation depends on values far away from the point being studied, the behavior of the solution at a distance can influence the local smoothness. The paper carefully separates these distant influences from the local behavior, proving that the distant effects do not disrupt the smoothness established locally. The author shows that these distant influences can be controlled and bounded, allowing the local smoothness to propagate throughout the entire system. This separation of local and global effects is a key innovation, as it prevents the long-range interactions from overwhelming the delicate local structure.

The proof relies on a technique called a compactness argument, which essentially says that if you have a sequence of solutions that are getting closer and closer to a limit, that limit must also be a solution. The author constructs a sequence of approximations and shows that they converge to a smooth function. This convergence is not guaranteed in the singular case, so the author had to develop new tools to ensure it happens. These tools involve measuring the "energy" of the solution in a way that accounts for the singular nature of the equations. By showing that this energy behaves in a predictable way, the author ensures that the sequence of approximations does not break down but instead settles into a smooth limit.

The result is a partial answer to a long-standing problem. The paper proves that for a specific range of parameters, the solutions are smooth. It does not claim to solve the problem for all possible parameters, but it covers the most difficult case where the equations are singular and the fractional order is close to one. This is a substantial step forward, as it confirms that even in the most challenging scenarios, the underlying mathematics preserves a fundamental property of smoothness. The work provides a framework that other researchers can build upon, potentially extending the result to other ranges of parameters or different types of equations.

In the broader context of mathematical physics, this finding reinforces the idea that nature tends toward smoothness, even when the rules governing it are complex and nonlocal. It suggests that the irregularities we might expect in such systems are, in fact, an illusion that disappears when viewed with the right mathematical tools. The paper does not just offer a new theorem; it offers a new way of seeing these equations, showing that their complexity can be tamed through careful analysis and the right perspective. This is a quiet but powerful contribution to the field, adding a piece to the puzzle of how we understand the smoothness of the world around us.

Technical Summary: Towards Gradient Hölder Regularity for Singular Fractional p-Laplace Equations

Problem Statement
The paper addresses the interior regularity of bounded weak solutions to the singular fractional $p$ -Laplace equation:
$(-\Delta_p)^s u = 0 \quad \text{in } \Omega \subset \mathbb{R}^n,$
where $n \ge 2$ , $0 < s < 1$ , and $1 < p < 2$ . The operator is defined via the principal value integral involving the singular flux $J_p(t) = |t|^{p-2}t$ . While regularity results such as boundedness, Hölder continuity, and Harnack inequalities are well-established for this equation, the interior $C^{1,\alpha}$ regularity (gradient Hölder continuity) remains an open problem in the singular regime ( $1 < p < 2$ ). Previous results for the superquadratic case ( $p \ge 2$ ) or for operators with globally Lipschitz constitutive functions do not apply here due to the lack of uniform ellipticity after affine normalization and the singularity of the flux near the origin.

Methodology and Proof Strategy
The author employs a compactness and improvement-of-flatness approach, adapted specifically to the singular nonlocal setting. The proof relies on several novel technical components:

Intrinsic Affine Normalization:
Unlike standard excess-decay arguments, the paper introduces a "slope-normalized" framework. The remainder $w = u - \ell$ (where $\ell$ is an affine map with slope $q$ ) is scaled by an intrinsic amplitude $A$ . This leads to a normalized slope $Q = q/A$ and a normalized remainder $v$ . The energy is measured using a shifted Bregman energy:
$D\Phi(Q \cdot (x-y); \delta v(x,y)),$
where $\Phi(t) = |t|^p/p$ . This Bregman remainder naturally captures two regimes: quadratic behavior when the remainder is small relative to the slope ( $|b| \ll |a|$ ) and $p$ -growth when the remainder dominates ( $|b| \gtrsim |a|$ ). This allows the treatment of both bounded and diverging slopes within a single framework.
Slope-Uniform Compactness:
A central difficulty is that the limit of the equation as $s \uparrow 1$ depends on the behavior of the normalized slope $Q$ .
- If $Q_j \to Q \in \mathbb{R}^n$ , the limit is a shifted local $p$ -harmonic equation.
- If $|Q_j| \to \infty$ with $Q_j/|Q_j| \to e$ , the limit is a uniformly elliptic constant-coefficient quadratic equation (linearized around the large slope).
  The paper establishes a slope-uniform compactness theorem (Proposition 3.8) using radial compactification of the slope space. This ensures that the sequence of solutions converges to a local minimizer of the appropriate limit functional in both regimes, without an energy defect.
Exterior Flux and Tail Control:
Subtracting an affine map in a nonlocal equation alters the exterior interaction (the "tail"). The paper distinguishes between:
- Averaged tails: Required for the compactness argument.
- One-sided adverse tails: Arising in De Giorgi truncations (upper and lower).
  The author introduces a joint propagation mechanism (Section 7) where the exterior flux is controlled via annular $L^p$ estimates upgraded to $L^r$ estimates via interpolation. Crucially, the dependence of the tail bound on the local maximum is strictly sublinear (power $a = 1 - p/r < 1$ ). This sublinearity allows the iteration to close even when the tail is large, provided the structural condition $sp > p-1$ holds.
Intrinsic Excess and Iteration:
The proof constructs an intrinsic excess functional that is scale-invariant and compatible with the singular geometry. A "one-step improvement" lemma (Proposition 6.3) transfers the local affine decay from the compact limit equations back to the fractional equation, provided the exterior tails are controlled. This is iterated via a Campanato scheme to establish geometric decay of the affine approximations.

Key Contributions and Results
The main result is Theorem 1.1, which states:
Let $n \ge 2$ , $1 < p < 2$ , and let $\alpha_{loc}(n, p)$ be an admissible interior Hölder exponent for gradients of local $p$ -harmonic functions. For every $0 < \alpha < \alpha_{loc}(n, p)$ , there exists a threshold $s^* = s^*(n, p, \alpha) < 1$ such that for all $s \in (s^*, 1)$ , every bounded weak solution $u$ of $(-\Delta_p)^s u = 0$ in $B_2$ belongs to $C^{1,\alpha}(B_{1/2})$ .

The estimate is quantitative:
$\|\nabla u\|_{L^\infty(B_{1/2})} + [\nabla u]_{C^\alpha(B_{1/2})} \le C \|u\|_{L^\infty(\mathbb{R}^n)}.$

Significance and Scope

Partial Answer to an Open Problem: The paper provides a partial answer to the long-standing open $C^{1,\alpha}$ problem for singular fractional $p$ -Laplace equations. It establishes the result specifically in the "near-local" regime where $s$ is sufficiently close to 1.
Structural Range: The tail propagation argument works throughout the structural range $sp > p-1$. The restriction $s > s^*$ is used only in the compactness step to ensure the limit equation is local (either $p$ -harmonic or linear elliptic). If $s$ were far from 1, the limit would remain nonlocal, requiring different oscillation estimates not provided here.
Uniformity: The constants in the estimate are uniform as $s \uparrow 1$ , making the result quantitatively consistent with the local $p$ -harmonic gradient theory.
Novel Techniques: The introduction of the slope-normalized Bregman energy and the separation of averaged versus one-sided tails to achieve sublinear bootstrap closure are presented as the primary technical innovations enabling the proof.

The paper concludes that while the full range $0 < s < 1$ remains open for the singular case, the near-local limit has been resolved, bridging the gap between nonlocal fractional equations and their local counterparts in the singular regime.

More like this