Balayage, equilibrium measure, and Deny’s principle of positivity of mass for
α · 2024
Read Balayage, equilibrium measure, and Deny’s principle of positivity of mass for by α (2024) 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-2411-01221.txt.
Description
A 2024 arXiv research paper by α. In the theory of g_{\alpha} -potentials on a domain D\subset\mathbb{R}^{n} , n\geqslant 2 , g_{\alpha} being the \alpha -Green kernel associated with the \alpha -Riesz kernel |x-y|^{\alpha-n} of order \alpha\in(0,n) , \alpha\leqslant 2 , we establish the existence and uniqueness of the g_{\alpha} -balayage \mu^{F} of a positive Radon measure \mu onto a relatively closed set F\subset D , we analyze its alternative characterizations, and we provide necessary and/or sufficient conditions for \mu^{F}(D)=\mu(D) to hold, given in terms of the \alpha -harmonic measure
Catalog record
| CREATOR | α |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2411-01221
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Balayage, equilibrium measure, and Deny’s principle of positivity of mass for α Abstract Abstract. In the theory of g_{\alpha} -potentials on a domain D\subset\mathbb{R}^{n} , n\geqslant 2 , g_{\alpha} being the \alpha -Green kernel associated with the \alpha -Riesz kernel |x-y|^{\alpha-n} of order \alpha\in(0,n) , \alpha\leqslant 2 , we establish the existence and uniqueness of the g_{\alpha} -balayage \mu^{F} of a positive Radon measure \mu onto a relatively closed set F\subset D , we analyze its alternative characterizations, and we provide necessary and/or sufficient conditions for \mu^{F}(D)=\mu(D) to hold, given in terms of the \alpha -harmonic measure of suitable Borel subsets of \overline{\mathbb{R}^{n}} , the one-point compactification of \mathbb{R}^{n} . As a by-product, we find necessary and/or sufficient conditions for the existence of the g_{\alpha} -equilibrium measure \gamma_{F} , \gamma_{F} being understood in an extended sense where \gamma_{F}(D) might be infinite. We also discover quite a surprising version of Deny’s principle of positivity of mass for g_{\alpha} -potentials, thereby significantly improving a previous result by Fuglede and Zorii (Ann. Acad. Sci. Fenn. Math., 2018). The results thus obtained are sharp, which is illustrated by means of a number of examples. Some open questions are also posed. 0 0 0 2010 Mathematics Subject Classification: Primary 31C15. 0 0 0 Key words: \alpha -Green balayage; \alpha -Green equilibrium measure; \alpha -harmonic measure; \alpha -thinness of a set at infinity; a refined version of Deny’s principle of positivity of mass for \alpha -Green potentials. 1. General conventions and preliminaries Fix an (open, connected) domain D\subset\mathbb{R}^{n} , n\geqslant 2 , which might coincide with the whole \mathbb{R}^{n} , and a positive (Radon) measure \mu on \mathbb{R}^{n} that is concentrated on D . (Being concentrated on a Borel set B\subset\mathbb{R}^{n} means that \mu(B^{c})=0 , where B^{c}:=\mathbb{R}^{n}\setminus B , or equivalently that \mu=\mu|_{B} , where \mu|_{B} is the trace of \mu to B .) All such \mu form the convex cone, denoted by \mathfrak{M}^{+}(D) . If D=\mathbb{R}^{n} , we simply write \mathfrak{M}^{+}:=\mathfrak{M}^{+}(\mathbb{R}^{n}) . Any element of \mathfrak{M}^{+}(D) can obviously be thought of as a positive (Radon) measure on D , where D is now treated as a locally compact (Hausdorff) space. For the theory of measures on a locally compact space, we refer to Bourbaki [ 7 ] or Edwards [ 12 ] . However, not every measure \xi on the locally compact space D can be extended to a measure on \mathbb{R}^{n} (unless, of course, \xi is bounded , i.e. \xi(D)<\infty ). Namely, \mathfrak{M}^{+}(D) consists of all positive measures on the locally compact space D that are extendible by 0 outside D to all of \mathbb{R}^{n} ; or equivalently, \mathfrak{M}^{+}(D) is the class of all \nu|_{D} , \nu ranging over \mathfrak{M}^{+} . We use one and the same symbol for \mu\in\mathfrak{M}^{+}(D) interpreted either as a measure on the locally compact space D , or as the trace \nu|_{D} of some \nu\in\mathfrak{M}^{+} to D . The extension of \mu\in\mathfrak{M}^{+}(D) by 0 outside D to \mathbb{R}^{n} will be denoted by \mu as well. Fix \alpha\in(0,n) , \alpha\leqslant 2 . It follows from the above that for \mu\in\mathfrak{M}^{+}(D) , one can apply the theory of potentials with respect to both the \alpha -Riesz and \alpha -Green kernels, denoted by \kappa_{\alpha} and g_{\alpha} , respectively. That is, \kappa_{\alpha}(x,y):=|x-y|^{\alpha-n} , x,y\in\mathbb{R}^{n} , and g_{\alpha}(x,y):=g^{\alpha}_{D}(x,y):=U_{\kappa_{\alpha}}^{\varepsilon_{x}}(y)-U_{\kappa_{\alpha}}^{(\varepsilon_{x})_{\kappa_{\alpha}}^{Y}}(y),\quad(x,y)\in D\times\mathbb{R}^{n}, (1.1) where U_{\kappa_{\alpha}}^{\nu}(\cdot):=\int\kappa_{\alpha}(\cdot,z)\,d\nu(z) is the \kappa_{\alpha} -potential of \nu\in\mathfrak{M}^{+} , \varepsilon_{x} is the unit Dirac measure at x\in D , and (\varepsilon_{x})_{\kappa_{\alpha}}^{Y} is the \kappa_{\alpha} -balayage of \varepsilon_{x} onto the closed (in \mathbb{R}^{n} ) set Y:=D^{c} , uniquely characterized within \mathfrak{M}^{+}(Y) by means of the equality U_{\kappa_{\alpha}}^{(\varepsilon_{x})_{\kappa_{\alpha}}^{Y}}=U_{\kappa_{\alpha}}^{\varepsilon_{x}}\quad\text{quasi-everywhere on $Y$}. (1.2) (For more details, see e.g. [ 13 , 15 , 17 , 18 ] . Recall that quasi-everywhere ( q.e. ) means that the equality holds on all of Y , except for a subset of outer \kappa_{\alpha} -capacity zero. For the theory of inner and outer \kappa_{\alpha} -capacities, we refer to Landkof [ 17 , Section II.2] . It should however be stressed here that all the sets, appearing in the current work, are Borel , so that their inner and outer \kappa_{\alpha} -capacities coincide [ 17 , Theorem 2.8] .) When speaking of \nu\in\mathfrak{M}^{+} , we always understand that U_{\kappa_{\alpha}}^{\nu} is not identically infinite on \mathbb{R}^{n} , which according to [ 17 , Section I.3.7] occurs if and only if \int_{|y|>1}\frac{d\nu(y)}{|y|^{n-\alpha}}<\infty. (1.3) Then, actually, U_{\kappa_{\alpha}}^{\nu} is finite q.e. on \mathbb{R}^{n} [ 17 , Section III.1.1] . Note that ( 1.3 ) necessarily holds if \nu is either bounded, or of finite \kappa_{\alpha} -energy I_{\kappa_{\alpha}}(\nu) (cf. ( 1.4 ) with D:=\mathbb{R}^{n} ). This work is to continue the investigation of the theory of g_{\alpha} -potentials, initiated by Riesz [ 18 ] and Frostman [ 13 ] and proceeded further by Fuglede and Zorii [ 15 ] . As in [ 13 , 15 , 18 ] , in the present paper we deal only with analytic aspects of this theory, leaving aside its probabilistic counterpart, cf. [ 2 , 11 ] . Let \mathcal{E}^{+}_{g_{\alpha}} stand for the cone of all positive measures \xi on D of finite g_{\alpha} -energy : I_{g_{\alpha}}(\xi):=\int_{D\times D}g_{\alpha}(x,y)\,d(\xi\otimes\xi)(x,y)<\infty. (1.4) The main fact discovered in [ 15 ] (see Theorem 4.11 therein) is that the kernel g_{\alpha} is perfect , which means that the cone \mathcal{E}^{+}_{g_{\alpha}} is complete in the strong topology, introduced by means of the energy norm \|\xi\|_{g_{\alpha}}:=\sqrt{I_{g_{\alpha}}(\xi)} , and that the strong topology on \mathcal{E}^{+}_{g_{\alpha}} is finer than the vague topology on \mathcal{E}^{+}_{g_{\alpha}} , the topology of pointwise convergence on the class C_{0}(D) of all continuous functions \varphi:D\to\mathbb{R} of compact (in D ) support. 1 1 1 For D=\mathbb{R}^{n} , this goes back to Cartan [ 8 ] (if \alpha=2 ) and Deny [ 9 ] (if \alpha\in(0,2] ). With regard to the concept of a perfect kernel on a locally compact space, see Fuglede [ 14 , Section 3.3] . Another useful fact discovered in [ 15 ] (see Theorem 4.6 therein) is that the kernel g_{\alpha} satisfies the domination principle in the following quite strong form: for any \xi\in\mathcal{E}^{+}_{g_{\alpha}} , any (extendible) \mu\in\mathfrak{M}^{+}(D) , and any positive \alpha -superharmonic function \psi on \mathbb{R}^{n} such that U^{\xi}_{g_{\alpha}}\leqslant U^{\mu}_{g_{\alpha}}+\psi \xi -a.e., the same inequality holds true on all of D . 2 2 2 For the concept of \alpha -superharmonic function on \mathbb{R}^{n} , see [ 17 , Chapter I] . Here U_{g_{\alpha}}^{\xi}(\cdot):=\int g_{\alpha}(\cdot,y)\,d\xi(y) is the g_{\alpha} -potential of a measure \xi on D . The present work is concerned with the theory of g_{\alpha} -balayage and some relevant questions, see Sections 2 – 5 for the results obtained and Section 6 for their proofs. The analysis performed is mainly based on the above-mentioned results from [ 15 ] , the theory of \kappa_{\alpha} -balayage [ 2 , 15 , 17 , 20 , 21 ] , and the theory of balayage on a locally compact space with respect to a perfect kernel satisfying suitable maximum principles, the latter being developed in the author’s recent papers [ 22 ] – [ 24 ] and [ 26 ] – [ 28 ] . Let c_{\kappa_{\alpha}}(B) , resp. c_{g_{\alpha}}(B) , denote the capacity of a Borel set B\subset\mathbb{R}^{n} , resp. B\subset D , with respect to the kernel \kappa_{\alpha} ,