Composition operators and Rational Inner Functions on the bidisc: A geometric approach

Athanasios Beslikas · 2026

Read Composition operators and Rational Inner Functions on the bidisc: A geometric approach by Athanasios Beslikas (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-2604-20330.txt.

Introduction

Beslikas studies composition operators on weighted Bergman spaces over the bidisc, induced by rational inner functions or a rational inner function plus a smooth function. His main result gives one uniform criterion for every beta from just above minus one to zero: boundedness matches transversal intersection of level sets for non-smooth symbols. Tangential intersections on the singularity, when they occur, must be of high order. The criterion extends a Bayart-Kosiński characterization to non-smooth self maps of the bidisc, using Clark measures and Puiseux factorizations.

Catalog record

CREATORAthanasios Beslikas
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE90426
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.20330

Browse the full library · About the Tower of Babel

Full text

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

Composition operators and Rational Inner Functions on the bidisc: A geometric approach Athanasios Beslikas Abstract . We study composition operators acting on the weighted Bergman spaces on the bidisc, i.e. C Φ : A β 2 ​ ( 𝔻 2 ) → A β 2 ​ ( 𝔻 2 ) C_{\Phi}:A^{2}_{\beta}(\mathbb{D}^{2})\to A^{2}_{\beta}(\mathbb{D}^{2}) where Φ \Phi is induced by rational inner functions (RIFs) or a RIF and a smooth function (mixed case). Our approach is geometric. Our main result is a uniform criterion for all β ∈ ( − 1 , 0 ] \beta\in(-1,0] that can be summarized as follows: Boundedness of the composition operator is equivalent to transversal intersection of the level sets for non-smooth symbols, under the assumption that if any tangential intersection occurs on the singularity it must be of high order. This extends the characterization of Bayart-Kosiński to the non-smooth self maps of the bidisc. To reach our conclusions, we utilize results obtained by Anderson, Bergqvist, Bickel, Cima and Sola on Clark measures associated to RIFs and Puiseux factorizations. 2020 Mathematics Subject Classification Primary 32A08, 32A37 1. Introduction Let \mathbb{D}^{2}:=\bigg\{z=(z_{1},z_{2})\in\mathbb{C}^{2}:|z_{1}|<1,|z_{2}|<1\bigg\} denote the unit bidisc in two complex variables and \mathbb{T}^{2}:=\bigg\{\zeta=(\zeta_{1},\zeta_{2})\in\mathbb{C}^{2}:|\zeta_{1}|=1,|\zeta_{2}|=1\bigg\} denote the bitorus, the distinguished boundary of the bidisc. Consider f\in\mathcal{O}(\mathbb{D}^{2},\mathbb{C}), a holomorphic function on the bidisc, and \Phi\in\mathcal{O}(\mathbb{D}^{2},\mathbb{D}^{2}) a holomorphic self map of the said domain. The composition operator C_{\Phi}(f):=(f\circ\Phi)(z_{1},z_{2}),\quad(z_{1},z_{2})\in\mathbb{D}^{2}, defines a new holomorphic function on the bidisc. Let us for a moment be more abstract. The first problem that one studies from a functional-analytic point of view is when this operator defines a bounded linear operator on a Banach or Hilbert space X, i.e. for what self maps \varphi of the domain \Omega\subset\mathbb{C}^{n}, the inequality \|C_{\varphi}(f)\|_{X}\leq C\|f\|_{X}, is satisfied. In the case where X=H^{2}(\mathbb{D)} or X=A^{2}_{\beta}(\mathbb{D}) the Hardy and (weighted) Bergman spaces of the unit disc, the problem is resolved via the celebrated Littlewood subordination principle ; all self maps \varphi\in\mathcal{O}(\mathbb{D},\mathbb{D}) induce bounded composition operators in both spaces. Inspired by the work of Shapiro (see the book [ 32 ] for a comprehensive study of composition operators in one complex variable), who built the geometric foundation in the study of composition operators on the one-dimensional setting, this work pursues a geometric interpretation of the problem in two complex variables. In one variable, understanding the geometry of the self map \varphi\in\mathcal{O}(\mathbb{D},\mathbb{D}), leads to results on compactness of the composition operator in Hardy and Bergman spaces. As an example that highlights this geometrical aspect of the problem in one complex variable, Shapiro states in his seminal book [ 32 ] , page 33, the following intuitive description of compactness of the composition operator; “ C_{\varphi} is compact if and only if \varphi(z) does not get too close to the unit circle too often.” More recently, mathematicians have turned their attention to the polydisc \mathbb{D}^{n} and the action of the composition operator on Banach and Hilbert spaces defined on this domain. The problem of characterizing the symbols \Phi that induce bounded composition operator in the Hardy and Bergman spaces of the bidisc was first considered in the work of Koo, Stessin and Zhu [ 26 ] . Later, Bayart partially corrected the previous results and his paper [ 5 ] together with the more recent paper of Kosiński [ 28 ] , led to a characterization of \mathcal{C}^{2}(\overline{\mathbb{D}}^{2})- smooth self maps that induce bounded composition operators on the Hardy and Bergman space of the bidisc. This characterization is often reffered to as the first-order condition . All of the studies are technical in nature, a testament of the difficulty of the problem in several complex variables. The higher dimensional setting restricts our intuition on the geometry of the self maps of the bidisc and how these mappings behave when their values approach the bitorus. In this paper, our goal is to study the self maps of the bidisc with singular behavior, appearing in the papers [ 9 , 8 ] , but this time we focus on the geometry of the self maps. Specifically, for each coordinate function we will picture its level sets on \mathbb{T}^{2}, compare them, and collect results based on this comparison. Our goal is to relate our results to the first-order condition. For convenience of the reader, let us state this result. Theorem 1.1 . (First-order condition) Let \Phi\in\mathcal{O}(\mathbb{D}^{2},\mathbb{D}^{2})\cap\mathcal{C}^{2}(\overline{\mathbb{D}^{2}}) . Then the composition operator C_{\Phi}:A^{2}_{\beta}(\mathbb{D}^{2})\to A^{2}_{\beta}(\mathbb{D}^{2}) is bounded, if and only if for all \zeta\in\mathbb{T}^{2} such that \Phi(\zeta)\in\mathbb{T}^{2}, the derivative d_{\zeta}\Phi is invertible. One could reformulate the first-order condition geometrically as follows (see Proposition 3.7 for a proof of this fact). Proposition 1.2 . (First-order condition is equivalent to transversality) Let \Phi=(\phi,\psi)\in\mathcal{O}(\mathbb{D}^{2},\mathbb{D}^{2})\cap\mathcal{C}^{2}(\overline{\mathbb{D}^{2}}), and consider all \zeta\in\mathbb{T}^{2} such that \Phi(\zeta)\in\mathbb{T}^{2}. Then the following are equivalent: (1) The derivative d_{\zeta}\Phi is invertible for all \zeta\in\mathbb{T}^{2} such that \Phi(\zeta)\in\mathbb{T}^{2} . (2) The boundary level curves \mathcal{C}_{\phi(\zeta)}(\phi)\quad\mathrm{and}\quad\mathcal{C}_{\psi(\zeta)}(\psi) are smooth for all \zeta\in\mathbb{T}^{2} such that \Phi(\zeta)\in\mathbb{T}^{2} and intersect transversely there. (3) C_{\Phi}:A^{2}_{\beta}(\mathbb{D}^{2})\to A^{2}_{\beta}(\mathbb{D}^{2}) is bounded. Here, by transversality we mean that the curves intersect at a strictly positive angle. The target of the present study is to prove a geometric principle for boundedness of the composition operator in the spirit of the compactness criterion of Shapiro, that is “ The composition operator C_{\Phi} is bounded on the Bergman space of the bidisc, if and only if the coordinate functions touch the bitorus transversally”. This, as we saw, is true for smooth symbols, but does this result extend to non-smooth cases? 1.1. Structure of the paper In the next two subsections 1.2 and 1.3 of the introduction, we provide the setting in which we will study our problem. In Section 2 , we state explicitly our main results that contribute in the study of composition operators. In Section 3 , we provide the necessary lemmata-theorems-propositions that we shall invoke in the duration of the paper. We split this section in 3 smaller ones for convenience of the reader. Then, we provide the proofs of the main results in Section 4 . In Section 5 we present four concrete examples that illustrate several of our main results in detail. 1.2. Bergman spaces on the bidisc and Carleson measures Among the most natural spaces of holomorphic functions are the (weighted) Bergman spaces in a domain. Our focus will be the polydisc \mathbb{D}^{n}, and particularly the bidisc \mathbb{D}^{2} . For \beta\geq-1, the weighted Bergman spaces are A^{2}_{\beta}(\mathbb{D}^{2}):=\Bigg\{f\in\mathcal{O}(\mathbb{D}^{2}):\int_{\mathbb{D}^{2}}|f(z_{1},z_{2})|^{2}(1-|z_{1}|^{2})^{\beta}(1-|z_{2}|^{2})^{\beta}dV(z_{1},z_{2})<+\infty\Bigg\}. Here, dV(z_{1},z_{2})=dA(z_{1})\times dA(z_{2}), where dA(z)=\frac{1}{\pi}dxdy,\quad z=x+iy\in\mathbb{D}, denotes the normalized Lebesgue measure on the unit disc. For the sake of notational clarity, dV_{\beta} will denote the radially weighted volume measure. Whenever \beta=-1 we recover the Hardy spaces A^{2}_{-1}(\mathbb{D}^{2})=H^{2}(\mathbb{D}^{2}), and for \beta=0 we obtain the classical Bergman space A^{2}_{0}(\mathbb{D}^{2})=A^{2}(\mathbb{D}^{2}). The main tool in the paper is the characterization of Carleson measures for the weighted Bergman spaces. Carleson measures for the radially weighted Bergman spaces have been fully characterized, thus providing us with a sufficient and necessary condition for the boundedness of the composition operator. Before we state the related Lemma, let us recall some definitions. Let \tilde{\delta}=(\delta_{1},\delta_{2})\in(0,1)^{2} and \zeta=(\zeta_{1},\zeta_{2})\in\mathbb{T}^{2}. The two-dimensional Carleson box is the set S(\zeta,\tilde{\delta})=\{z\in\mathbb{D}^{2}:|z_{j}-\zeta_{j}|<\delta_{j},\quad j=1,2\}. The characterization of the Carleson measures that can be found in [ 12 ] gives us for free a straightforward criterion for the boundedness of the composition operator C_{\Phi} on

Read the complete text