🔢 mathematics

Local regularity for nonlocal equations with variable exponents

This paper establishes the local boundedness and Hölder continuity of weak solutions to nonlocal variational equations with variable exponents, demonstrating that continuity and boundedness of the exponent suffice for boundedness, while log-Hölder-type conditions both inside and outside the domain are required for Hölder regularity.

Original authors: Jamil Chaker, Minhyun Kim

Published 2026-08-28

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jamil Chaker, Minhyun Kim

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 world of mathematics, scientists often study how things change and smooth out over space. Imagine a landscape where the rules for how a surface bends or how heat spreads are not the same everywhere. In some spots, the material might be stiff and unyielding, while in others, it is soft and pliable. This is the realm of variable exponents, a field where the mathematical "rules" of a system shift from point to point. For decades, mathematicians have understood how these systems behave when they are local, meaning the behavior at one spot depends only on its immediate neighbors. However, many real-world phenomena, from the movement of particles in a fluid to the spread of information in a network, are nonlocal. In these cases, what happens at a single point can be influenced by conditions far away, across the entire system. Understanding how these complex, shifting systems settle into a stable state is a major challenge, because the usual mathematical tools often break down when the rules change and the influence stretches across long distances.

A team of researchers has now taken a significant step forward in understanding these nonlocal systems with shifting rules. They focused on a specific type of mathematical problem that describes how a system minimizes its energy, a process that leads to a stable, smooth configuration. The researchers wanted to know if the solutions to these problems are well-behaved. Specifically, they asked two questions: first, are the solutions bounded, meaning they do not shoot off to infinity or collapse into nothingness? Second, are the solutions smooth enough to be described as continuous, without sudden, jagged jumps? Their work confirms that under reasonable conditions, the answers are yes. They proved that solutions to these complex equations are always locally bounded, provided the changing rules of the system do not fluctuate too wildly. Furthermore, they showed that if the rules change in a sufficiently controlled manner, the solutions are not just bounded but also smooth, possessing a specific type of continuity that allows them to be predicted with high precision.

To reach these conclusions, the authors had to navigate a landscape where the mathematical "exponent," which dictates the system's behavior, varies across space. In simpler, local problems, mathematicians have long known that if this exponent changes too erratically, the system can develop singularities or fail to behave smoothly. The researchers found that a similar constraint is necessary for these nonlocal problems. They identified a condition where the exponent must not change too quickly over small distances, a requirement that ensures the system retains enough regularity to be analyzed. However, because these problems are nonlocal, the researchers discovered a new, unique requirement. The rules governing the system outside the area of interest also matter. They proved that the behavior of the exponent in the region surrounding the main domain must be compatible with the behavior inside it. If the rules outside are too different from the rules inside, the smoothness of the solution can be compromised. This finding highlights a fundamental difference between local and nonlocal systems: in the nonlocal world, the environment surrounding a problem is just as critical as the problem itself.

The team developed a new set of mathematical tools to handle these variable conditions. They created a method to estimate the size of the solutions, effectively proving that they cannot grow infinitely large. This was achieved by adapting a classic technique known as iteration, which involves refining an estimate step by step until a precise bound is reached. In this nonlocal setting, the process was complicated by the fact that the exponent changes at every step, requiring the researchers to carefully balance different mathematical powers to ensure the estimates held true. Once they established that the solutions were bounded, they moved on to proving their smoothness. They demonstrated that if the exponent satisfies the necessary continuity conditions both inside and outside the domain, the solutions will not have sharp corners or breaks. Instead, they will vary gradually, maintaining a consistent texture throughout the space.

The significance of this work lies in its ability to unify the understanding of systems that are both variable and nonlocal. By proving that these systems produce stable and smooth outcomes, the researchers have provided a solid foundation for future studies in physics, engineering, and other fields where such complex interactions occur. Their results show that even when the underlying rules of a system are not constant, and even when distant parts of the system influence each other, nature still tends toward order and predictability. The findings do not rely on simulations or approximations but are rigorous mathematical proofs. They confirm that the chaotic potential of variable, long-range interactions is tamed by the right conditions, ensuring that the solutions remain well-behaved and understandable. This advances the broader goal of mathematical physics: to find the hidden order in systems that appear, at first glance, to be too complex to decipher.

Technical Summary: Local Regularity for Nonlocal Equations with Variable Exponents

Problem Statement
This paper investigates the local regularity properties of minimizers for nonlocal variational functionals with variable exponents and the corresponding weak solutions to their Euler–Lagrange equations. The primary functional under consideration is:
$\mathcal{F}(u) = \int_{\mathbb{R}^n} \int_{\mathbb{R}^n} \frac{|u(x) - u(y)|^{p(x,y)}}{p(x,y)|x - y|^{n+sp(x,y)}} \, dy \, dx$
where $n \in \mathbb{N}$ , $s \in (0, 1)$ , and $p: \mathbb{R}^n \times \mathbb{R}^n \to \mathbb{R}$ is a continuous, symmetric function satisfying $1 < \inf p \le \sup p < +\infty$ . The associated Euler–Lagrange equation is the fractional $p(\cdot, \cdot)$ -Laplacian:
$(-\Delta)^s_{p(\cdot,\cdot)} u = 0 \quad \text{in } \Omega.$
The authors aim to establish local boundedness and local Hölder continuity for weak solutions, addressing the specific challenges introduced by the variable exponent $p(x,y)$ in a nonlocal setting. Unlike local problems where the exponent depends only on $x$ , the nonlocal nature requires $p$ to be defined on $\mathbb{R}^n \times \mathbb{R}^n$ , introducing dependencies between points inside and outside the domain.

Methodology
The authors employ a De Giorgi iteration technique adapted for nonlocal functionals with variable exponents. The methodology relies on several key components:

Caccioppoli-type Estimates: The authors derive an improved Caccioppoli-type estimate for weak subsolutions and supersolutions. This estimate generalizes previous results for constant exponents (specifically those by Cozzi) to the variable exponent case. A significant technical difficulty arises because the variable exponent leads to terms involving both $p^+$ and $p^-$ in the iteration, creating a mismatch that complicates the control of the supremum.
Fractional Sobolev Embeddings: To handle the iteration, the paper utilizes fractional Sobolev embeddings. While variable exponent Sobolev embeddings exist, the authors opt for embeddings with constant exponents (Theorem 2.4) combined with a specific lemma (Lemma 4.3) that establishes a continuous embedding $W^{s,p(\cdot,\cdot)}(\Omega) \hookrightarrow W^{\sigma,q}(\Omega)$ for $\sigma < s$ and $q < p^-$ . This approach avoids the need for stronger assumptions on $p$ required by direct variable exponent embeddings.
Growth Lemma and Isoperimetric Inequalities: To prove Hölder continuity, the authors establish a growth lemma. This requires an improved Caccioppoli estimate and a fractional De Giorgi isoperimetric-type inequality. The proof of the growth lemma involves a careful analysis of the "nonlocal tail" terms, which are sensitive to the behavior of $p$ outside the domain.
Iterative Schemes: The proofs of both local boundedness and Hölder continuity utilize recursive sequences (De Giorgi iteration) to show that the measure of sets where the solution exceeds certain levels decays to zero, implying boundedness and continuity.

Key Assumptions
The paper introduces two specific conditions on the exponent $p$ to ensure regularity:

(P1) Log-Hölder-type condition inside the domain: For any ball $B \subset \Omega$ , the oscillation of $p$ on $B \times B$ must satisfy $R^{p^-(B \times B) - p^+(B \times B)} \le L$ . This is analogous to the log-Hölder continuity required for local variable exponent problems.
(P2) Condition outside the domain: For any ball $B \subset \Omega$ , the exponent must satisfy $p^+(B \times B^c) \le p^+(B \times B)$ and $p^-(B \times B^c) \le p^-(B \times B)$ . This assumption is crucial for controlling the nonlocal tail terms where the interaction occurs between points inside the domain and points outside.

Main Results

Local Boundedness (Theorem 1.4):
The authors prove that weak solutions to the equation are locally bounded in $\Omega$ , provided the solution satisfies a tail condition (1.8) regarding its behavior at infinity. Notably, this result holds assuming only that $p$ is continuous and bounded; it does not require the log-Hölder-type conditions (P1) or (P2). This extends previous results for constant exponents and improves upon recent work by Ho and Kim, which required log-Hölder assumptions even for boundedness in subcritical cases.
Local Hölder Continuity (Theorem 1.5):
The authors prove that bounded weak solutions are locally Hölder continuous in $\Omega$ . This result does require the additional assumptions (P1) and (P2).
- Significance of (P1): It resolves the mismatch of exponents ( $p^+$ vs. $p^-$ ) that arises during the De Giorgi iteration when analyzing the modulus of continuity.
- Significance of (P2): It ensures that the variable exponent in the nonlocal tail term does not disrupt the iteration process, a difficulty unique to nonlocal problems with variable exponents that does not appear in local variable exponent problems or nonlocal constant exponent problems.

Significance and Claims
The paper claims to establish the foundational regularity theory for nonlocal variational problems with variable exponents. The authors highlight that:

The condition (P1) is shown to be weaker than the standard log-Hölder continuity of $p$ as a $2n$ -variable function, yet sufficient for regularity.
The necessity of condition (P2) is a novel finding specific to the nonlocal setting, addressing the interaction between the exponent inside and outside the domain.
The results are obtained without requiring the variable exponent Sobolev embedding with variable exponents, instead using a refined embedding into spaces with lower differentiability and integrability (Lemma 4.3).
The work provides quantitative estimates for the supremum of weak subsolutions (Theorem 4.1) and establishes the growth lemma (Lemma 5.2) as the core mechanism for Hölder regularity.

The authors acknowledge that similar results have been studied by Jihoon Ok using a different approach, but emphasize that their work provides a distinct derivation and explicitly addresses the specific structural assumptions (P1) and (P2) required for the nonlocal variable exponent case.

More like this