Sobolev regularity for the nonlocal ( 1 , p ) (1,p) -Laplace equations in the superquadratic case
Dingding Li and Chao Zhang ∗ · 2025
Read Sobolev regularity for the nonlocal ( 1 , p ) (1,p) -Laplace equations in the superquadratic case by Dingding Li and Chao Zhang ∗ (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-23288.txt.
Introduction
Dingding Li and Chao Zhang prove regularity results for solutions of a nonlocal equation that mixes two growth regimes, a one-growth and a p-growth, in the harder superquadratic case. Such equations model materials whose response changes with the size of the gradient, and nonlocal versions let distant points interact directly. Good behavior of weak solutions is not automatic.
The authors establish interior Sobolev regularity, meaning solutions have well-behaved derivatives inside the domain. As a byproduct they obtain explicit Hölder continuity estimates, with constants that can in principle be written down. Their proof uses finite difference quotients, careful tail estimates for the nonlocal terms, refined energy bounds, and a Moser iteration to climb from rough to smooth.
Catalog record
| CREATOR | Dingding Li and Chao Zhang ∗ |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 91 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-23288
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Sobolev regularity for the nonlocal ( 1 , p ) (1,p) -Laplace equations in the superquadratic case Thanks: ∗ Corresponding author. Dingding Li and Chao Zhang ∗ Abstract . We investigate the interior Sobolev regularity of weak solutions to the nonlocal ( 1 , p ) (1,p) -Laplace equations in the superquadratic case p ≥ 2 p\geq 2 . More precisely, for the fractional differentiability index s p ∈ ( 0 , p − 1 p ] s_{p}\in\left(0,\frac{p-1}{p}\right] , we establish that the weak solution u u exhibits W loc γ , q W_{\rm loc}^{\gamma,q} -regularity for any γ ∈ ( 0 , s p p p − 1 ) \gamma\in\left(0,\frac{s_{p}p}{p-1}\right) and q ≥ p q\geq p . When s p ∈ ( p − 1 p , 1 ) s_{p}\in\left(\frac{p-1}{p},1\right) , the gradient ∇ u \nabla u of the weak solution is shown to exist and belong to ( L loc q ( Ω ) ) N \left(L^{q}_{\mathrm{loc}}(\Omega)\right)^{N} . As a product, the explicit Hölder continuity estimates of weak solutions are derived. The proof relies on a detailed analysis of the structural characteristics of ( 1 , p ) (1,p) -growth in the nonlocal setting, combined with the finite difference quotient method, tail estimates, refined energy estimates, and a Moser-type iteration scheme. 2020 Mathematics Subject Classification 35B65, 35D30, 35J60, 35R11 1. Introduction Let \Omega be a bounded domain in \mathbb{R}^{N} with N\geq 2 . In this paper, we consider the following nonlocal (1,p) -Laplace equation \displaystyle(-\Delta_{1})^{s_{1}}u+(-\Delta_{p})^{s_{p}}u=0\quad\text{in }\Omega, (1.1) where s_{1},s_{p}\in(0,1) , p\geq 2 and the fractional operators (-\Delta_{1})^{s_{1}} and (-\Delta_{p})^{s_{p}} are defined by \displaystyle(-\Delta_{1})^{s_{1}}u(x):=2\mathrm{P.V.}\int_{\mathbb{R}^{N}}\frac{u(x)-u(y)}{|u(x)-u(y)|}\frac{dy}{|x-y|^{N+s_{1}}} and \displaystyle(-\Delta_{p})^{s_{p}}u(x):=2\mathrm{P.V.}\int_{\mathbb{R}^{N}}\frac{|u(x)-u(y)|^{p-2}(u(x)-u(y))}{|x-y|^{N+s_{p}p}}\,dy for x\in\Omega . Here \mathrm{P.V.} denotes the integral taken in the principal value sense. The equations of the type considered in ( 1.1 ) were recently proposed by Górny, Mazón and Toledo [ 38 ] in a general setting of random walk spaces. These equations incorporate two random walk structures so that the associated functionals have different growth on each structure, which can be regarded as a special case of (p,q) -growth problems. Since the pioneering works [ 44 , 45 ] , this class of problems has been extensively investigated. We refer the readers to [ 56 , 12 , 4 , 18 , 46 ] and the references therein for detailed overviews. 1.1. Overview of related results. Let us start with the problems with local 1 -growth. For the inhomogeneous 1 -Laplace Dirichlet problem \displaystyle\begin{cases}-\Delta_{1}u:=-\mathrm{div}\left(\frac{\nabla u}{|\nabla u|}\right)=f&\textmd{in }\Omega,\\ u=0&\textmd{on }\partial\Omega,\end{cases} (1.2) it is well-known that the diffusion singularity of the 1-Laplacian \Delta_{1} manifests prominently on the set \left\{Du=0\right\} , commonly referred to as a facet. To address this issue, a vector field Z\in L^{\infty}(\Omega;\mathbb{R}^{N}) was introduced by using the Anzellotti-Frid-Chen pairing theory (see [ 3 , 13 , 14 , 15 ] ) that serves as a generalized representative of the quotient \frac{Du}{|Du|} imprecisely. We refer to a series of works by Mazón and collaborators [ 47 , 40 , 49 , 50 ] for more comprehensive treatments of this approach. Moreover, problem ( 1.2 ) is usually addressed as the limit version for p -Laplace equations as p goes to 1^{+} . The behavior of solutions to ( 1.2 ) has been rigorously analyzed under various assumptions on the right-hand side term f , see [ 16 , 42 , 53 ] . In these works, a smallness condition on f is typically imposed to ensure that the solution remains finite almost everywhere in \Omega . However, within the L^{1} framework, one cannot generally expect that the solutions are finite almost everywhere in \Omega . In this context, Mercaldo, Segura de León and Trombetti [ 54 ] established the existence of renormalized solutions. Regarding the regularity of classical 1-growth problems, as far as we know, since the ellipticity of 1-Laplacian becomes singular over the facet \left\{Du=0\right\} , the solutions of -\Delta_{1}u=f may fail to be continuously differentiable. Nevertheless, the results on problems with nearly linear growth stand in sharp contrast to those in the 1-growth regime. In particular, De Filippis and Mingione [ 21 ] established the Schauder-type estimates for minimizers of the functional: \displaystyle\mathcal{L}(w,\Omega):=\int_{\Omega}\left[c(x)|Dw|\log(1+|Dw|)+a(x)|Dw|^{q}\right]\,dx,\quad q>1. Under suitable conditions on the coefficients a(x) and c(x) , these results yield C^{1,\alpha} -regularity for the corresponding minimizers. We also refer to [ 19 , 23 , 32 , 33 , 30 , 57 ] for more results on the problems with nearly linear growth. When it comes to the nonlocal counterpart \displaystyle(-\Delta_{1})^{s}u=0, there remain relatively few results addressing problems with 1-growth. It is worth noting that Mazón et al. have made a significant contribution to this field. In their seminal works [ 51 , 48 ] , they not only introduced a rigorous definition of weak solutions for nonlocal evolution problems with 1-growth but also characterized the new qualitative phenomena inherent in these problems. In contrast to the local case, where the expression \frac{Du}{|Du|} arises naturally, the nonlocal counterpart takes the form \frac{u(x)-u(y)}{|u(x)-u(y)|} . To properly interpret this quantity, one must identify a function Z representing the nonlocal quotient. In this direction, Andreu et al. [ 1 , 2 ] demonstrated that for a class of nonlocal diffusion equations, the corresponding quantity satisfies Z(x,y,t)\in\mathrm{sgn}(u(x,t)-u(y,t)) , where \mathrm{sgn}(\cdot) denotes the set-valued sign function with \mathrm{sgn}(0)=[-1,1] (see also [ 52 ] ). Bucur et al. [ 10 ] investigated the existence of (s,1) -harmonic function and established their equivalence to the minimizers of functional \displaystyle\int_{\mathbb{R}^{N}}\int_{\mathbb{R}^{N}}\frac{|u(x)-u(y)|}{|x-y|^{N+s}}\,dxdy. Assuming the inhomogeneous term f belongs to L^{\frac{N}{s}}(\Omega) with a smallness assumption, Bucur [ 9 ] studied the minimizer of \mathcal{F}^{s_{p}}_{p} given by \displaystyle\mathcal{F}^{s_{p}}_{p}(u):=\frac{1}{2p}\int_{\mathbb{R}^{N}}\int_{\mathbb{R}^{N}}\frac{|u(x)-u(y)|^{p}}{|x-y|^{N+s_{p}p}}\,dxdy-\int_{\Omega}fu\,dx, where p\geq 1 and s_{p}=N+s-\frac{N}{p}\in[s,1) . And it is shown that the minimizers correspond to weak solutions of the nonlocal 1-Laplacian problem and a “flatness” result was obtained, which implies that the expression of Z towards \frac{u(x)-u(y)}{|u(x)-u(y)|} may not precise. Under certain natural structural conditions, Bucur et al. [ 11 ] proved that any s -minimal function u lies in C(\Omega)\cap L^{\infty}(\Omega) and admits a continuous extension \overline{u}\in C(\overline{\Omega}) . For further regularity results, we refer to the work of Novaga and Onoue [ 58 ] , who investigated the minimizers of a nonlocal variational problem arising from an image denoising model: \displaystyle\text{min}\left\{\mathcal{F}_{K,f}(u)\big|u\in BV_{K}(\mathbb{R}^{N})\cap L^{2}(\mathbb{R}^{N})\right\}, where \mathcal{F}_{K,f} is defined by \displaystyle\mathcal{F}_{K,f}(u):=\frac{1}{2}\int_{\mathbb{R}^{N}\times\mathbb{R}^{N}}K(|x-y|)|u(x)-u(y)|dxdy+\frac{1}{2}\int_{\mathbb{R}^{N}}\left(u(x)-f(x)\right)^{2}dx and the space BV_{K}(\mathbb{R}^{N}) is given by \displaystyle BV_{K}(\mathbb{R}^{N}):=\left\{u\in L^{1}(\mathbb{R}^{N})\bigg|\int_{\mathbb{R}^{N}\times\mathbb{R}^{N}}K(|x-y|)|u(x)-u(y)|dxdy<+\infty\right\} with K being a kernel singular at the origin. In two dimensions, under the assumption that the original image f is locally \beta -Hölder continuous with \beta\in(1-s,1] , the authors showed that the minimizers preserve the same local Hölder regularity. Finally, we would like to turn to the perturbed 1-Laplace equation of the form \displaystyle-\mathrm{div}\left(\frac{\nabla u}{|\nabla u|}\right)-\mathrm{div}\left(|\nabla u|^{p-2}\nabla u\right)=f\quad\text{in }\Omega. For the cases p=2 and p=3 , this equation arises in fluid mechanics [ 29 ] and materials science [ 59 ] respectively. Such problems have been investigated only in a limited number of specific settings. The anisotropic case involving the p -Laplacian when it has linear growth on some of the coordinates has also been studied, for instance, in [ 37 , 55 ] . Employing De Giorgi-Nash-Moser theory, Tsubouchi [ 60 ] proved the local Lipschitz continuity of weak solutions. Under the additional assumption of convexity for weak solutions, Giga