Riesz α \alpha -capacity of Cantor sets and cyclicity in Dirichlet-type spaces
Dimitrios Vavitsas, Jujie Wu, Konstantinos Zarvalis · 2026
Read Riesz α \alpha -capacity of Cantor sets and cyclicity in Dirichlet-type spaces by Dimitrios Vavitsas, Jujie Wu, Konstantinos Zarvalis (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-2604-10324.txt.
Introduction
Vavitsas, Wu, and Zarvalis examine the cycling threshold for functions in Dirichlet-type spaces. For a fixed key index, they construct a smooth function that is cycling for every index below the key index but fails to be cycling at the key index itself. The function serves as a counter-case to survival of cycling at the key index. The build uses broad Cantor sets and studies their Riesz size, linking cycling ways to the shape of these sets.
Catalog record
| CREATOR | Dimitrios Vavitsas, Jujie Wu, Konstantinos Zarvalis |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67302 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.10324
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Riesz α \alpha -capacity of Cantor sets and cyclicity in Dirichlet-type spaces Dimitrios Vavitsas, Jujie Wu, Konstantinos Zarvalis Abstract . We examine the threshold of the cyclicity for functions in Dirichlet-type spaces 𝒟 α \mathcal{D}_{\alpha} , α ∈ ( 0 , 1 ] \alpha\in(0,1] . Given a fixed α ∗ ∈ ( 0 , 1 ] \alpha^{*}\in(0,1] , we construct a holomorphic function f ∈ D α ∗ f\in D_{\alpha^{*}} which is cyclic in 𝒟 α \mathcal{D}_{\alpha} for all α < α ∗ \alpha<\alpha^{*} , but fails to be cyclic in 𝒟 α ∗ \mathcal{D}_{\alpha^{*}} . This function serves as a counterexample to the persistence of cyclicity at the critical index α ∗ \alpha^{*} . Throughout the construction process, we work with generalized Cantor sets and study their Riesz α \alpha -capacity. 2020 Mathematics Subject Classification Primary: 30C85, 46E20, 47A16; Secondary: 30H99, 31A15 1. Introduction This article explores the cyclicity of functions with respect to the Dirichlet-type spaces \mathcal{D}_{\alpha} , \alpha\in(0,1] , of the unit disc {\mathbb{D}} in the complex plane {\mathbb{C}} . Our objective is to discover whether cyclicity in these spaces is “closed” in the following sense: given a fixed \alpha^{*}\in(0,1] , does the cyclicity of f in {\mathcal{D}}_{\alpha} , for all \alpha<\alpha^{*} , imply the cyclicity of f in {\mathcal{D}}_{\alpha^{*}} ? We will answer this question in the negative by constructing explicit counterexamples. In order to produce our counterexamples, we shall work beforehand with generalized Cantor sets of the unit circle {\mathbb{T}}\vcentcolon=\partial{\mathbb{D}} . By the term generalized Cantor set, we mean Cantor sets produced as follows: Take the unit circle and remove an open subarc with -1 in the middle, so that two closed arcs of equal length with common endpoint 1 remain. Then, remove two open subarcs from their middles, so that four closed arcs remain (two of them have common endpoint 1 ). Continue this process inductively, and what remains is a generalized Cantor set in {\mathbb{T}} . Such sets are known to be Carleson sets. We are going to study generalized Cantor sets with regard to their Riesz \alpha -capacity , \alpha\in(0,1] (the case \alpha=1 yields the well-known logarithmic capacity ). Riesz \alpha -capacity is a tool that gauges the size of subsets of the unit circle, but essentially differs from the Lebesgue measure. As a matter of fact, it provides a more delicate approach in measuring sets, since there exist compact subsets of the unit circle whose Lebesgue measure is zero, yet they are of positive Riesz \alpha -capacity for certain parameters \alpha\in(0,1] . Our first objective is to make explicit constructions of generalized Cantor sets satisfying prescribed properties with respect to their Riesz capacities. More specifically, given \alpha^{*}\in(0,1] , we will create such a set E_{\alpha^{*}} whose Riesz \alpha -capacity vanishes for all \alpha\in(0,\alpha^{*}) , whereas its \alpha^{*} -Riesz capacity is positive. In addition, whenever \alpha^{*}\in(0,1) we will proceed to a similar construction of a set F_{\alpha^{*}} whose Riesz \alpha -capacity vanishes if and only if \alpha\in(0,\alpha^{*}] . Explicit constructions of such Carleson sets are of interest in their own right. However, due to the strong ties betweeen potential theoretic tools and functional analysis, we may utilize our generalized Cantor sets to produce results relative to spaces of holomorphic functions. As a matter of fact, the notion of Riesz \alpha -capacity is closely related to the cyclicity of functions in the Dirichlet-type spaces of the unit disc. This important topic has been at the forefront of many advancements, from the seminal paper [ 7 ] of Brown and Shields, to a wide variety of high-quality research conducted in the past decades; see [ 6 , 12 , 13 , 14 , 15 ] to name a few. The Dirichlet-type spaces {\mathcal{D}}_{\alpha} , \alpha\in{\mathbb{R}} , form a family consisting of Hilbert spaces of holomorphic functions in the unit disc. For the values \alpha>1 , the space {\mathcal{D}}_{\alpha} is an algebra. On the other hand, whenever \alpha\leq 1 , \mathcal{D}_{\alpha} is a Banach space which is not an algebra, and its cyclic vectors play a role similar to that played by invertible elements in algebras. Classical Hilbert spaces such as the Bergman, the Hardy and the Dirichlet spaces are all special instances of Dirichlet-type spaces; in particular, in terms of equivalent norms, {\mathcal{D}}_{-1} is the Bergman space, {\mathcal{D}}_{0} is the Hardy space, and {\mathcal{D}}_{1} is the Dirichlet space (denoted by {\mathcal{D}} ). For a presentation of the theory concerning this family of weighted Dirichlet spaces, we refer the interested reader to [ 1 ] , [ 7 ] and [ 12 , Chapter 1] . More information and concrete definitions follow in Section 2 . We will mainly deal with Dirichlet-type spaces \mathcal{D}_{\alpha} where the parameter lies in the interval (0,1] . These spaces provide a natural scale of Banach spaces linking the classical Hardy space to the classical Dirichlet space. Let us point out that in the Hardy space ( \alpha=0 ), a complete characterization of cyclicity was achieved by Beurling; see [ 4 ] . In fact, a function f is cyclic in \mathcal{D}_{0} if and only if it is outer. The characterization of cyclic in {\mathcal{D}}_{\alpha} functions, for \alpha\in(0,1] , remains an open problem. It can be easily deduced that if \alpha_{1},\alpha_{2}\in{\mathbb{R}} with \alpha_{1}<\alpha_{2} and f\in\mathcal{D}_{\alpha_{2}} is cyclic in \mathcal{D}_{\alpha_{2}} , then it necessarily belongs to and is cyclic in \mathcal{D}_{\alpha_{1}} as well. The exact converse trivially does not hold. So, it is natural to wonder whether the following statement holds: Problem . Fix \alpha^{*}\in(0,1] and suppose that f is cyclic in \mathcal{D}_{\alpha} , for all \alpha<\alpha^{*} . Is f necessarily cyclic in \mathcal{D}_{\alpha^{*}} ? This question constitutes the main motivation of our work. We are going to answer it in the negative by providing two different explicit counterexamples. Given a function f as in the statement above, let us denote its zero set by \mathcal{Z}(f)=\{z\in{\mathbb{D}}\vcentcolon f(z)=0\} . A wide variety of published work points to the fact that the cyclicity of f in the spaces \mathcal{D}_{\alpha} is inextricably related to the size of \mathcal{Z}(f^{*}) with respect to its Riesz \alpha -capacity, where f^{*} denotes the radial limit function of f . Recall that by Fatou’s Theorem, f^{*} is defined almost everywhere on {\mathbb{T}} . With this in mind, we will construct, as we mentioned, a generalized Cantor set E_{\alpha^{*}}\subset{\mathbb{T}} that is a Carleson set and whose Riesz \alpha -capacity vanishes for all \alpha\in(0,\alpha^{*}) , whereas its Riesz \alpha^{*} -capacity is positive. This will in turn lead to the construction of two fundamentally different functions whose zero set coincides with E_{\alpha^{*}} and which act as our desired counterexamples. The first “threshold” function is constructed exclusively through the fact that E_{\alpha^{*}} is a Carleson set and is independent of the actual values of its Riesz capacities. Surprisingly, the resulting function f belongs to {\mathcal{D}}_{\alpha^{*}} , can be created to be as smooth as desired on the closed unit disc, but will still not be cyclic in {\mathcal{D}}_{\alpha^{*}} . An important takeway is that cyclicity is not a “closed” notion even when the function we investigate satisfies nice differentiability conditions. On the other hand, the second construction relies profoundly on the vanishing of the Riesz \alpha -capacity of E_{\alpha^{*}} for \alpha\in(0,\alpha^{*}) . In