Two-Scale Frostman Measures

Nicolás Angelini, Úrsula Molter · 2025

Read Two-Scale Frostman Measures by Nicolás Angelini, Úrsula Molter (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-2511-04302.txt.

Introduction

Nicolas Angelini and Ursula Molter build a two-scale Frostman measure framework linking classical Hausdorff dimension with the intermediate dimensions introduced by Falconer, Fraser, and Kempton. These middle dimensions probe how a set's size behaves between the finest and box-counting scales, and Frostman measures are the standard tool for certifying dimension from below.

A new geometric value is defined for a set. Under mild assumptions, a family of measures on the set meets two decay conditions at once, one matching Hausdorff and one matching intermediate Frostman inequalities. Those two-scale measures then pin down the set's dimension across both regimes in a single construction, joining tools that were used apart before.

Catalog record

CREATORNicolás Angelini, Úrsula Molter
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE18 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-04302

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Two-Scale Frostman Measures Nicolás Angelini, Úrsula Molter Abstract . We establish a unified Frostman-type framework connecting the classical Hausdorff dimension with the family of intermediate dimensions dim θ \dim_{\theta} recently introduced by Falconer, Fraser and Kempton. We define a new geometric quantity 𝒟 ⁡ ( E ) \mathcal{D}(E) and prove that, under mild assumptions, there exists a family of measures { μ δ } \{\mu_{\delta}\} supported on E E satisfying two simultaneous decay conditions, corresponding to the Hausdorff and intermediate Frostman inequalities. Such ( δ , s , t ) (\delta,s,t) -Frostman measures allow for a two-scale characterization of the dimension of E E . 2020 Mathematics Subject Classification 28A80, 28A75 1. Introduction The study of fractal dimensions provides a quantitative framework for describing the fine-scale geometry of sets across multiple scales. A unifying theme in this area is the characterization of dimension through the decay of measures supported on the set. The classical example of this correspondence is Frostman’s lemma, which relates the Hausdorff dimension of a closed set E\subset\mathbb{R}^{n} to the polynomial decay of measures supported on it. Specifically, Frostman [ 11 ] proved that \mathcal{H}^{s}(E)>0 if and only if there exists a Borel probability measure \mu supported on E such that (1) \mu(B(x,r))\leq c\,r^{s}\quad\text{for all }x\in\mathbb{R}^{n}\text{ and }r>0, for some constant c>0 . Here, \mathcal{H}^{s} denotes the s -dimensional Hausdorff measure. More recently, Falconer, Fraser, and Kempton [ 8 ] introduced a one-parameter family of intermediate dimensions , denoted by the lower and upper quantities \underline{\dim}_{\theta}E and \overline{\dim}_{\theta}E for \theta\in[0,1] . These dimensions interpolate continuously between the Hausdorff and box-counting dimensions, capturing a broader range of scaling behaviors by restricting the admissible diameters in the definition of the Hausdorff dimension to lie within a geometric range determined by \theta . The study of intermediate dimensions has developed rapidly, and several recent works have explored their geometric and measure-theoretic properties. For instance, projection theorems have been investigated in [ 6 , 2 ] , attainable forms in [ 4 , 3 ] , and images of more general functions in [ 7 ] . In parallel, various generalizations and alternative definitions of measures have been proposed [ 5 , 1 ] . In analogy with the classical Hausdorff dimension, intermediate dimensions can also be characterized in terms of the polynomial decay of measures supported on E . The following proposition provides a Frostman-type characterization for these dimensions. Proposition 1.1 . Let E\subset\mathbb{R}^{n} be compact and \theta\in(0,1] . (1) If \underline{\dim}_{\theta}E>0 , then for every s\in(0,\underline{\dim}_{\theta}E) there exists a constant c>0 such that, for every \delta\in(0,1) , one can find a Borel probability measure \mu_{\delta} supported on E satisfying \mu_{\delta}(B(x,r))\leq c\,r^{s}\quad\text{for all }x\in E\text{ and }\delta^{1/\theta}\leq r\leq\delta. (2) If \overline{\dim}_{\theta}E>0 , then for every s\in(0,\overline{\dim}_{\theta}E) there exists a constant c>0 such that, for every \delta_{0}>0 , there exist \delta\in(0,\delta_{0}) and a Borel probability measure \mu_{\delta} supported on E satisfying (2) \mu_{\delta}(B(x,r))\leq c\,r^{s}\quad\text{for all }x\in E\text{ and }\delta^{1/\theta}\leq r\leq\delta. In this paper, we introduce a new geometric quantity \mathcal{D}(E) and establish sufficient conditions ensuring the existence of a family of measures \{\mu_{\delta}\} that simultaneously satisfy both decay estimates ( 1 ) and ( 2 ). This provides a unified framework linking the classical Frostman lemma with its intermediate counterpart. Theorem . Let \theta>0 and E\subset\mathbb{R}^{d} be a compact set such that 0<\mathcal{D}(E) . Then for all 0<t<\overline{\mbox{\rm dim}}_{\,\theta}E and all 0<s<\mathcal{D}(E) with s\leq t , there exists c>0 such that for all \delta_{0}>0 there exist \delta\in(0,\delta_{0}) and a Radon measure \mu_{\delta} supported on E satisfying (3) \mu_{\delta}(B(x,r))\leq\begin{cases}c\,(\delta^{1/\theta})^{t-s}r^{s},&\text{if }r\in(0,\delta^{1/\theta}),\\[4.0pt] c\,r^{t},&\text{if }r\in[\delta^{1/\theta},\delta].\end{cases} for all x\in E . Equivalently, if 0<t<\underline{\mbox{\rm dim}}_{\,\theta}E and 0<s<\mathcal{D}(E) , there exist constants c>0 and \delta_{0}>0 such that for all \delta\in(0,\delta_{0}] there exists a measure \mu_{\delta} supported on E satisfying ( 3 ). A Borel probability measure satisfying ( 3 ) will be called a (\delta,s,t) -Frostman measure. 2. Preliminaries Throughout the whole document \mathcal{D}_{n} will denote the classical dyadic partition of \mathbb{R}^{d} into 2^{dn} half-open disjoint cubes of diameter \sqrt{d}\,2^{-n} . B(x,r) , r>0 will denote the open ball in \mathbb{R}^{d} with center x and radio r and \mathcal{B} the Borel subsets of \mathbb{R}^{d} . Given a non-empty set A\subset\mathbb{R}^{d} we will write |A| for the diameter of the set and the distance between A and a point x will be d(x,A)=\displaystyle\inf_{y\in A}|x-y| . The Hausdorff dimension of A will be \dim_{\rm H}A and \dim_{B}A will denote Box-counting dimension of A . The reader can refer to [ 9 ] for more information on these dimensions. \mathcal{H}^{s} will always refer to the Hausdorff s- measure and \mathcal{L}^{d} the d- dimensional Lebesgue measure. The support of a measure \mu on \mathbb{R}^{d} is defined as spt\,\mu=\mathbb{R}^{d}\setminus\{x:\exists\,r>0\mbox{ such that }\mu(B(x,r))=0\} . Throughout the paper Radon measure will mean a locally finite and Borel regular measure. This document concerns the \theta intermediate dimensions, which were introduced in [ 8 ] and are defined as follows. Definition 2.1 . Let F\subseteq\mathbb{R}^{d} be bounded. For 0\leq\theta\leq 1 we define the lower \theta -intermediate dimension of F by \displaystyle\underline{\mbox{\rm dim}}_{\,\theta}F=\inf\big\{ \displaystyle s\geq 0:\ \forall\ \varepsilon>0,\ {\rm and\ all}\ \delta_{0}>0,{\rm there}\ \exists\ 0<\delta\leq\delta_{0},\ {\rm and}\ \displaystyle\{U_{i}\}_{i\in I}\ :F\subseteq\cup_{i\in I}U_{i}\ :\delta^{1/\theta}\leq|U_{i}|\leq\delta\ {\rm and}\ \sum_{i\in I}|U_{i}|^{s}\leq\varepsilon\big\}. Similarly, we define the upper \theta -intermediate dimension of F by \displaystyle\overline{\mbox{\rm dim}}_{\,\theta}F=\inf\big\{ \displaystyle s\geq 0:\ \forall\ \varepsilon>0,\ {\rm there}\ \exists\ \delta_{0}>0,\ :\ \forall\ 0<\delta\leq\delta_{0},\mbox{ \rm there }\ \exists \displaystyle\{U_{i}\}_{i\in I}\ :F\subseteq\cup_{i\in I}U_{i}\ :\delta^{1/\theta}\leq|U_{i}|\leq\delta\ {\rm and}\ \sum_{i\in I}|U_{i}|^{s}\leq\varepsilon\big\}. For all \theta\in(0,1] and a compact set A\in\mathbb{R}^{d} we have \dim_{\rm H}A\leq\overline{\mbox{\rm dim}}_{\,\theta}A\leq\overline{\mbox{\rm dim}}_{\rm B}\,A and similarly with the lower case replacing \overline{\mbox{\rm dim}}_{\rm B}\, with \underline{\mbox{\rm dim}}_{\rm B}\, . This spectrum of dimensions has the property of being continuous for \theta\in(0,1] , leaving the natural question of the continuity in \theta=0 as a problem to study. The reader can find more information in [ 8 ] . Finally, the lower dimension of a non-empty set E\subset\mathbb{R}^{d} is defined as \displaystyle\dim_{L}E=\sup\Big\{ \displaystyle\alpha:\text{ there are constants }c,\rho>0\text{ such that} \displaystyle\inf_{x\in E}N_{r}(B(x,R)\cap E)\geq c\left(\frac{R}{r}\right)^{\alpha}\text{ for all }0<r<R<\rho\Big\}. For a compact set F\subset\mathbb{R}^{d} we have 0\leq\dim_{L}F\leq\dim_{\rm H}F\leq\underline{\mbox{\rm dim}}_{\,\theta}F\leq\overline{\mbox{\rm dim}}_{\,\theta}F\leq d. The lower dimension is sensitive to the local structure of the set, for instance if F has an isolated point then \dim_{L}F=0 and if F is self-similar set then \dim_{L}F=\dim_{H}F . For more information on that dimension we refer to the Fraser’s book [ 10 ] . 3. Existence of (\delta,s,t)- Frostman measures Our main interest is to find conditions on the set E such that there exist a family of measures \{\mu_{\delta}\} that simultaneously satisfy both decay estimates ( 1 ) and ( 2 ). In order to be able to construct them, we first need to introduce a new parameter, which gives us information about the distribution of the set when refining dyadic cubes. Recall that \#A is the number of elements of the set A . Definition 3.1 . We define the Dyadic dimension as follows: \displaystyle\mathcal{D}(E):=\liminf_{n\to\infty}\frac{\log\mathcal{N}_{n}(E)}{\log 2} where \mathcal{N}_{n}(E)=\displaystyle\min_{Q\in\mathcal{D}_{n}}\#\{Q^{\prime}\in\mathcal{D}_{n+1}:\ Q^{\prime}\cap(E\cap Q)\neq\emptyset\} . Theorem 3.2 . Let \theta>0 and E\subset\mathbb{R}^{d} be a compact set, such that 0<\mathcal{D}(E) . Then for all 0<t<\overline{\mbox{\rm dim}}_{\,\theta}E and all 0<s<\mathcal{D}(E) with s\leq t , there exists c>0 such that for all \delta_{0}>0 exist \delta\in(0,\delta_{0}) and a Radon measure \mu_{\delta} , supported on E , satisfying (4) \mu_{\delta}(B(x,r))\leq\left\{\begin{array}[]{lcc}c(\delta^{1/\theta})^{t-s}r^{s}&if&r\in(0,\delta^{1/\theta})\\

Read the complete text