Examples of critically cyclic functions in the Dirichlet spaces of the ball

Pouriya Torkinejad Ziarati · 2026

Read Examples of critically cyclic functions in the Dirichlet spaces of the ball by Pouriya Torkinejad Ziarati (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-2601-08651.txt.

Introduction

Pouriya Torkinejad Ziarati constructs critically cyclic functions in Dirichlet spaces of the ball. The 2026 arXiv paper works in the Dirichlet space on the two-dimensional unit ball. The author builds holomorphic functions with a sharp cyclicity threshold. Each example has an index between one half and two such that the function is cyclic exactly up to that index. Cyclic functions can approximate every function in the space through shifts. Sharp examples map the border between cyclic and non-cyclic behavior for analysts studying these function spaces.

Catalog record

CREATORPouriya Torkinejad Ziarati
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE39 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-08651

Browse the full library · About the Tower of Babel

Full text

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

Examples of critically cyclic functions in the Dirichlet spaces of the ball Pouriya Torkinejad Ziarati Abstract . In this work, we construct examples of holomorphic functions in D 2 ​ ( 𝔹 2 ) D_{2}(\mathbb{B}_{2}) , the Dirichlet space on 𝔹 2 \mathbb{B}_{2} , for which there exists an index α c ∈ [ 1 2 , 2 ] \alpha_{c}\in[\frac{1}{2},2] such that the function is cyclic in D α ​ ( 𝔹 2 ) D_{\alpha}(\mathbb{B}_{2}) if and only if α ≤ α c \alpha\leq\alpha_{c} . To this end, we use the notion of interpolation sets in smooth ball algebras, as studied by Bruna, Ortega, Chaumat, and Chollet. 2020 Mathematics Subject Classification 32F45 1. Introduction Consider a Banach space of holomorphic functions \mathcal{A} defined on a domain \Omega in \mathbb{C}^{n} . A function f\in\mathcal{A} is called cyclic if the linear span of f multiplied by all multipliers of \mathcal{A} is dense in \mathcal{A} . (Here, m is a multiplier of \mathcal{A} if mf\in\mathcal{A} for all f\in\mathcal{A} .) Determining whether a function is cyclic is a space-dependent task. Roughly speaking, on a fixed domain, a more restrictive(larger) norm makes the conditions for cyclicity harder to satisfy. For instance, a function that is not cyclic in the Hardy space H^{2}(\mathbb{D}) is not cyclic in the Dirichlet space D_{1}(\mathbb{D}) either. Another important factor in the study of cyclicity is the zero set of the function. Indeed, let z_{0}\in\overline{\Omega} be such that the evaluation functional \Lambda_{z_{0}}:\mathcal{A}\to\mathbb{C},\quad f\mapsto f(z_{0}), is bounded. Then any function admitting a zero at z_{0} fails to be cyclic. In particular, this situation often occurs when z_{0}\in\Omega , which shifts the challenge to the study of boundary behavior of functions. In Dirichlet-type spaces on the unit ball D_{\alpha}(\mathbb{B}_{n}) , which are the subject of this study, one may define for any f\in D_{\alpha}(\mathbb{B}_{n}) a non-tangential boundary limit f^{*} outside a set of \alpha -capacity zero [ 5 ] . It turns out that, for f to be cyclic, the zero set of f^{*} , \mathcal{Z}(f^{*}) , cannot have positive \alpha -capacity. A converse to this fact, however, is more subtle and has been the subject of many studies in one and several variables. In one variable, in 1984, Brown and Shields conjectured that an outer function f\in D_{1}(\mathbb{D}) is cyclic if and only if \mathcal{Z}(f^{*}) has zero logarithmic capacity [ 2 ] . In 2009, El-Fallah, Kellay, and Ransford provided a partial answer for functions continuous up to the boundary with a specific type of zero set [ 10 ] , and later extended this result to Dirichlet-type spaces by replacing logarithmic capacity with \alpha -capacity [ 11 ] . Moreover, in harmonically weighted Dirichlet spaces, El-Fallah, Elmadani, and Labghail established a Brown–Shields type result under certain conditions on the weight measure [ 13 ] . In all of these results, the zero sets of A^{\infty}(\mathbb{D}) —the algebra of holomorphic functions that are \mathcal{C}^{\infty} -smooth up to the boundary—known as Beurling–Carleson sets, play an indispensable role. A modern treatment of these sets can be found in the work of Ivrii and Nicolau [ 15 ] . In several variables, on the unit ball, the concepts of interpolation sets and peak sets in smooth ball algebras (which are more restrictive than zero sets) may provide a useful tool for studying cyclicity. In [ 19 ] the author used a result of Chaumat and Chollet [ 6 ] on the peak sets of A^{\infty}(\mathbb{B}_{n}) to obtain sharp cyclicity conditions for a class of holomorphic functions defined on a neighborhood of the closed unit ball, whenever their zero set is a compact complex-tangential manifold (which, by [ 6 ] , is a peak set of A^{\infty}(\mathbb{B}_{n}) ), thereby providing further positive evidence for the conjecture posed in [ 20 ] . Consider a closed set E\subset\partial\mathbb{B}_{n} and \alpha_{c}\in[0,n] such that E has zero \alpha -capacity if and only if \alpha\leq\alpha_{c} . Then any function in D_{\alpha}(\mathbb{B}_{n}) whose zero set contains E is noncyclic whenever \alpha>\alpha_{c} . We call a function f\in D_{n}(\mathbb{B}_{n}) a critically cyclic function for the set E if it has E as its zero set and is cyclic in D_{\alpha}(\mathbb{B}_{n}) if and only if \alpha\leq\alpha_{c} . Here, we use a sufficiency result due to Bruna and Ortega [ 3 ] on the interpolation sets of A^{m,s}(\mathbb{B}_{n}) for 0<m+s<\infty and one provided by Chaumat and Chollet [ 8 ] in A^{\infty}(\mathbb{B}_{n}) to provide examples of critically cyclic functions in Dirichlet-type spaces on the unit ball. When these sets are contained in a transversal curve on the boundary of the unit ball, they satisfy a property called the \mathcal{K} -property. In the one-dimensional case, they are also known as porous sets and have been studied in [ 1 ] , [ 9 ] . As the results in [ 3 ] and [ 8 ] suggest, we study separately three interesting cases for the boundary zero set, namely the transversal , the complex tangential , and a case that is neither transversal nor complex tangential but is still totally real . Later in Section 2 , we will provide precise definitions of these cases. In our examples, which are for the unit ball in \mathbb{C}^{2} , one may observe that, depending on the case, the relationship between the Hausdorff dimension of the zero set and the critical index \alpha_{c} varies. In the transversal case, the zero set of our example has dimension d and \alpha_{c}=2-d . In the complex tangential case, however, we have \alpha_{c}=2-\frac{d}{2} . We also construct another example by taking a compact subset, of dimension d , of a transversal curve and extending it along a complex tangential direction, which yields a totally real set of dimension 1+d ; in this case, \alpha_{c}=\frac{3}{2}-d . In the transversal case, the following theorem gives us the necessary tool to prove our first main result: Proposition 1 . [ 3 , Theorem 4.3] Let E be a \mathcal{K} -set contained in a simple transverse curve \Gamma . Then there exists f\in A^{1}(\mathbb{B}_{n}) such that (1) |f(z)|\asymp d_{K}(z,E),\qquad z\in\bar{\mathbb{B}}_{n}. Moreover, f has this property that f^{k}\in A^{k}(\mathbb{B}_{n}) for k being a positive integer. Here, A^{k}(\mathbb{B}_{n})=\mathcal{O}(\mathbb{B}_{n})\cap\mathcal{C}^{k}(\bar{\mathbb{B}}_{n}) and, by d_{K}(z,E) we mean the Korányi pseudodistance between z\in\bar{\mathbb{B}}_{n} and the set E\subset\partial\mathbb{B}_{n} , as defined in Definition 6 . We call the function f in ( 1 ) an extremal function for the set E , and we establish the following result for it, which constitutes our first main result: Theorem 2 . Let f\in\mathcal{O}(\mathbb{B}_{2}) , E\subset\partial\mathbb{B}_{2} and \Gamma be as stated in Proposition 1 , moreover assume that whenever t\in\mathbb{R}^{+} is small enough we have: (2) |E_{t}|=|\{z\in\Gamma:d_{K}(z,E)\leq t\}|\asymp t^{1-d}, where |E_{t}| denotes the one dimensional Lebesgue measure of E_{t} , and moreover assume that d\in(0,1) is the Hausdorff dimension of E . Then, f is cyclic in D_{\alpha}(\mathbb{B}_{2}) if and only if \alpha\leq 2-d . In the complex tangential case, the following, which is a corollary of Theorem II in [ 3 ] will serve as the main tool to settle our second main result: Proposition 3 . Let M be a compact \mathcal{C}^{\infty} -smooth complex tangential curve and E\subseteq M be compact. Given 0<\epsilon<\frac{1}{2} , there exists h\in\mathcal{O}(\mathbb{B}_{2}) such that \mathcal{Z}(h)\cap\bar{\mathbb{B}}_{2}=E and for all k being

Read the complete text