Closing the gap: Maz’ya-Shaposhnikova and asymptotics of fractional perimeters
Elisa Davoli, Alberto Fanizza, Marco Picerni · 2026
Read Closing the gap: Maz’ya-Shaposhnikova and asymptotics of fractional perimeters by Elisa Davoli, Alberto Fanizza, Marco Picerni (2026) 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-2605-03955.txt.
Introduction
Elisa Davoli, Alberto Fanizza, and Marco Picerni generalize the Maz'ya-Shaposhnikova formula. It is a classical result. It describes what fractional Sobolev seminorms converge to as the fractional parameter tends to zero. The classical formula assumes functions in the square-integrable class, which forces vanishing at infinity. Many natural functions fail that assumption, leaving a gap the authors set out to close.
By introducing a notion of mass at infinity, they pin down the limit of Gagliardo seminorms localized on a bounded Lipschitz domain. They count only pairs with at least one point inside the domain. The limiting functional unifies the picture. It reduces to the ordinary norm for globally integrable functions. It recovers the pointwise limit of fractional perimeters for characteristic functions. It also encodes the limiting behavior in the sense of Gamma-convergence. A version for metric measure spaces follows.
Catalog record
| CREATOR | Elisa Davoli, Alberto Fanizza, Marco Picerni |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.03955
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Closing the gap: Maz’ya-Shaposhnikova and asymptotics of fractional perimeters Elisa Davoli, Alberto Fanizza, Marco Picerni Abstract . We prove a generalization of the Maz’ya-Shaposhnikova formula in the case p = 2 p=2 for functions that may not belong to L 2 ( ℝ d ) {L^{2}}(\mathbb{R}^{d}) and, thus, might not vanish at infinity. By introducing a notion of mass at infinity , we explicitly characterize the limit as s → 0 + s\to 0^{+} of Gagliardo seminorms localized on a bounded Lipschitz domain Ω \Omega . By ‘localized’, we mean here that we account only for interactions involving at least one point in Ω \Omega . The identified limiting functional provides a unifying framework to link the classical Maz’ya-Shaposhnikova formula and the asymptotics of nonlocal perimeters. On the one hand, it reduces to the classical L 2 L^{2} norm for functions that are globally integrable on ℝ d \mathbb{R}^{d} . On the other hand, it recovers the pointwise limit of s s -fractional perimeters when evaluated on characteristic functions of sets. We further show that the same functional encodes the asymptotic behavior of Gagliardo seminorms in the sense of Gamma-convergence with respect to the weak- L 2 L^{2} topology. Finally, we provide an extension to the setting of metric measure spaces. 2020 Mathematics Subject Classification 46E35, 26A33, 46B20 1. Introduction The study of the asymptotics of the Gagliardo seminorms is a central topic in nonlocal analysis, since it allows to provide a quantitative meaning to the idea that the Sobolev-Slobodeckij space W^{s,p} should be, in some sense, close to L^{p} when s is close to zero and close to W^{1,p} for s close to 1 . The first result in this direction is the Bourgain-Brezis-Mironescu (BBM) formula [ 3 , Theorem 2] , which deals with the case s\to 1^{-} . The formula states that for every function u\in L^{p}(\Omega) , with \Omega\subseteq\mathbb{R}^{d} bounded smooth domain, and for every p\in(1,+\infty) , (1.1) \lim_{s\to 1^{-}}(1-s)\iint_{\Omega\times\Omega}\frac{\left|u(x)-u(y)\right|^{p}}{\left|x-y\right|^{d+sp}}dxdy=K_{d,p}\|\nabla u\|_{L^{p}(\Omega)}^{p}, with the convention that the right hand side equals +\infty if u\notin W^{1,p}(\Omega) . Here, the constant K_{d,p} depends only on p and d . An analogous characterization for p=1 was also proved in [ 3 ] , but only under the a priori assumption that u\in W^{1,1}(\Omega) . This result was successfully extended to functions in BV(\Omega) by Dávila [ 9 , Theorem 1] (see also the development by Leoni and Spector [ 35 ] ). An analysis of the asymptotics in the sense of \Gamma -convergence was carried out by Ponce in [ 41 ] . Recently, the validity of the BBM formula has been investigated beyond the Euclidean setting, encompassing, for example, metric measure spaces satisfying a Poincaré inequality [ 15 , 39 , 38 ] , thin films [ 4 ] , as well as general anisotropic convolution kernels [ 26 , 10 , 11 , 33 ] . The Maz’ya-Shaposhnikova (MS) formula [ 36 ] provides a counterpart to ( 1.1 ) for the asymptotics of the s -Gagliardo seminorm as s approaches zero. To be precise, given a function u\in W^{s_{0},p}(\mathbb{R}^{d}) for some s_{0} and p\in(1,+\infty) , we have (1.2) \lim_{s\to 0^{+}}\frac{s}{2}\iint_{\mathbb{R}^{d}\times\mathbb{R}^{d}}\frac{\left|u(x)-u(y)\right|^{p}}{\left|x-y\right|^{d+sp}}dxdy=\frac{d\omega_{d}}{p}\|u\|_{L^{p}(\mathbb{R}^{d})}^{p}, where \omega_{d} is the Lebesgue measure of the unit ball in \mathbb{R}^{d} . We stress that ( 1.2 ) only holds if u\in L^{p}(\mathbb{R}^{d}) , and so only if u vanishes at infinity in a measure theoretic sense. The multiplicative factor d\omega_{d} , which is the d-1 Hausdorff measure of the unit sphere {\mathbb{S}^{d-1}} , accounts for the fact that, as s approaches zero, the fractional kernels \rho_{s}(x,y)=\frac{s}{\left|x-y\right|^{d+sp}} concentrate their mass at infinity (in a measure-theoretic sense). Note, also, that \omega_{d} encodes the asymptotic volume ratio (henceforth abbreviated AVR ) of the metric measure space (\mathbb{R}^{d},d_{E},\mathcal{L}^{d}) with respect to the standard volume function V(t)=t^{d} . The role played by the AVR was made much more evident by the works [ 29 , 28 , 27 ] , where the MS formula was generalized to the metric measure setting. Notable examples of spaces for which the formula holds are finite-dimensional (non-Euclidean) Banach spaces, Carnot Groups and MCP(K,N) spaces (which are metric spaces satisfying a synthetic upper bound on the dimension and a lower bound on the curvature, see [ 42 , 44 ] ). The choice of restricting the analysis in the classical MS formula to functions with global L^{p} -integrability on the whole \mathbb{R}^{d} comes with a structural limitation: forcing a function to vanish at infinity artificially suppresses its tail behavior, which the MS formula should intuitively emphasize. Roughly speaking, as s\to 0^{+} , the kernel s\left|x-y\right|^{-(d+sp)} decays very slowly at the infinity, causing long-range interactions to account for most of the energy. In fact, as s\to 0^{+} , the tail of the function \left|\cdot\right|^{-(d+sp)} approaches the threshold of non integrability, given by the exponent -d . From the point of view of the Gagliardo seminorm, this means that the interactions that prevail in this regime are the ones between points that lie “at infinite distance”. This suggests that, in the limit, we should see an interaction energy between points of \mathbb{R}^{d} and points “at infinity”. As an intuition, one can think that the fractional kernels force one of the two Lebesgue measures appearing in the integral ( 1.2 ) to become a d-1 -Hausdorff measure concentrated on a sphere representing all the possible (oriented) directions of \mathbb{R}^{d} . In the classical Maz’ya-Shaposhnikova formula, this phenomenon is completely hidden and the interactions at infinity disappear within the L^{p} norm. To properly overcome this limitation, i.e., to treat functions that may not vanish at infinity (and therefore fully capturing the highly nonlocal nature of the small- s regime), we consider the Gagliardo seminorm with interior-exterior interactions: (1.3) [u]_{W^{s,p}({Q_{\Omega}})}^{p}=\iint_{Q_{\Omega}}\frac{\left|u(x)-u(y)\right|^{p}}{\left|x-y\right|^{d+sp}}dxdy. Here, \Omega is a bounded, Lipschitz domain and {Q_{\Omega}}=(\mathbb{R}^{d}\times\mathbb{R}^{d})\setminus(\Omega^{c}\times\Omega^{c}) . We say that a measurable function u:\mathbb{R}^{d}\to\mathbb{R} belongs to W^{s,p}({Q_{\Omega}}) for some p\in[1,+\infty) if it belongs to L^{p}(\Omega) and [u]_{W^{s,p}({Q_{\Omega}})}<+\infty . Note that W^{s,p}({Q_{\Omega}})\subset L^{p}_{\text{loc}}(\mathbb{R}^{d}) . In fact, since \Omega is bounded, there exists a constant C=C(\Omega,s,d) such that, for every y\in\Omega and x\in\mathbb{R}^{d} , \frac{1}{|x-y|^{d+sp}}\geq C\frac{1}{1+|x|^{d+sp}}. Let now u\in W^{s,p}(Q_{\Omega}) and assume, without loss of generality, that u has zero mean in \Omega . Then, by Jensen’s inequality, \displaystyle[u]_{W^{s,p}({Q_{\Omega}})}^{p} \displaystyle\geq C\int_{\mathbb{R}^{d}}\frac{1}{1+|x|^{d+sp}}\int_{\Omega}|u(x)-u(y)|^{p}dydx \displaystyle\geq C\mathcal{L}^{d}(\Omega)\int_{\mathbb{R}^{d}}\frac{|u(x)|^{p}}{1+|x|^{d+sp}}dx. This extension of the Gagliardo seminorm, as evident from the decomposition (1.4) [u]_{W^{s,p}({Q_{\Omega}})}^{p}=[u]_{W^{s,p}(\Omega)}^{p}+2\int_{\Omega}\int_{\Omega^{c}}\frac{\left|u(x)-u(y)\right|^{p}}{\left|x-y\right|^{d+sp}}dxdy, isolates the energy arising from interactions involving at least one point in \Omega and ignores all the exterior-exterior contributions. Far from being a mere technical variant, this is the functional used to model nonlocal phenomena on bounded domains. Indeed, it is the natural energy for minimization problems with Dirichlet boundary conditions (where the value of the function needs to be prescribed on the whole \Omega^{c} , rather than just on \partial\Omega ) [ 14 , 25 , 40 ] . It also appears as singular perturbation in phase-transition models [ 43 ] , and in the theory of nonlocal minimal surfaces [ 16 , 17 , 18 , 19 , 21 ] . In particular it allows to define the fractional perimeter: \operatorname{Per}_{s}(E;\Omega)=\frac{1}{2}[\chi_{E}]_{W^{s,1}({Q_{\Omega}})}, which approximates the De Giorgi perimeter functional as s\to 1^{-} [ 1 , 23 ] . In the same spirit of the Maz’ya-Shaposhnikova formula, the asymptotic behavior of the