Skip to main content
Article Open Access

Characterizing the RIFs that induce bounded composition operators on the Bergman space of the bidisc

  • EMAIL logo
Published/Copyright: March 20, 2026

Abstract

In this short paper we will discuss recent advances on the problem of characterizing the boundedness of the composition operator acting on the Bergman spaces A β 2 ( D 2 ) whenever the self map Φ of the bidisc is induced by Rational Inner Functions. The problem stated here is submitted as part of the Problem List of the Young Researchers Workshop in Complex Analysis and Operator Theory “The Bench Math Session 2025”, organized in Jagiellonian University of Kraków at 10th–11th February 2025.

MSC 2020: 32A37; 32A40; 30J10

1 Introduction and formulation of the problem

Let.

A β 2 ( D 2 ) = f O ( D 2 , C ) : D 2 | f ( z 1 , z 2 ) | 2 1 | z 1 | 2 β 1 | z 2 | 2 β d V ( z 1 , z 2 ) < + ,

where β ≥ −1, denote the weighted Bergman spaces on the bidisc. Consider Φ = (ϕ, ψ), where Φ O ( D 2 , D 2 ) and assume that ϕ , ψ O ( D 2 , D ) are induced by Rational Inner Functions with singularities. Let p C [ z 1 , z 2 ] a polynomial in two complex variables, restricted on the bidisc. We assume that the bidegree of p satisfies deg p = ( n , m ) N 2 . We further assume that this polynomial has zeros on the bitorus T 2 . Then, RIFs on the bidisc D 2 are of the form ϕ = λ z 1 N z 2 M p ̃ p , where p ̃ ( z 1 , z 2 ) = z 1 n z 2 m p 1 z 1 ̄ , 1 z 2 ̄ ̄ , for z 1 , z 2 D 2 , and N, M are some positive integers. This characterization is due to Rudin, see [1]. As it will be pointed out in the references later on, it is evident that RIFs may posses singularities on the distinguished boundary T 2 , contrary to what happens on the one (complex) dimensional case. Kosiński and Bayart in [2], 3] respectively, studied the composition operator acting on the Hardy and Bergman space of the bidisc, whenever the self-map of the bidisc is induced by smooth symbols. In the paper [4], the author studied the case where the self map of the bidisc is induced by RIFs. Specifically, it was shown that whenever Φ = (ϕ, ϕ) and ϕ is a Rational Inner Function with a singularity, then C Φ : A 2 ( D 2 ) A 2 ( D 2 ) is not bounded. An example which showcases this, is the Rational Inner Mapping

Φ ( z 1 , z 2 ) = 2 z 1 z 2 z 1 z 2 2 z 1 z 2 , 2 z 1 z 2 z 1 z 2 2 z 1 z 2 ,

for ( z 1 , z 2 ) D 2 . In the sequel, we will provide an alternative proof for this example.

Contrary to the one-dimensional setting, where the only Rational Inner Functions are the Finite Blaschke Products, in two complex dimensions, Rational Inner Functions may have singularities. These singularities might occur on the distinguished boundary T 2 , as it is evident from the above example. A striking result of Greg Knese shows that Rational Inner Functions have non-tangential limit everywhere on T 2 , even on the singularities that might occur. The value of this non-tangential limit is unimodular. For more about Rational Inner Functions and their properties, like the membership of their derivatives on the Hardy and Dirichlet spaces of the bidisc, or the behavior of their zero sets, the reader can consult the articles [5], [6], [7], [8], [9], [10] and the references therein.

Having the essential background, let us now state the problem in question:

Problem 1.

Let Φ = (ϕ, ψ) be a self-map of the bidisc induced by RIFs with singularities, with ϕ, ψ not necessarily the same function. Give sufficient and/or necessary conditions depending on the nature of the RIFs such that the composition operator C Φ : A β 2 ( D 2 ) A β 2 ( D 2 ) is bounded or not bounded.

2 Recent results

In the papers [4] and [11] the problem of boundedness of the composition operator acting on the weighted Bergman spaces of the bidisc was introduced. Before we pass to the statements of the main results, let us briefly mention a Carleson measure criterium which is the main tool in the proofs of these Theorems.

Lemma 2.1.

Let Φ : D 2 D 2 be a holomorphic self-map of the bidisc. The composition operator C Φ : A a 2 ( D 2 ) A β 2 ( D 2 ) is bounded if and only if there is a constant C > 0 such that for every δ ̃ ( 0,2 ) 2 and ζ T 2 :

V β ( Φ 1 ( S ( ζ , δ ̃ ) ) ) C V a ( S ( ζ , δ ̃ ) ) .

Recall that a two-dimensional Carleson box is defined as.

S ( ζ , δ ̃ ) = ( z 1 , z 2 ) D 2 : | z 1 ζ 1 | < δ 1 , | z 2 ζ 2 | < δ 2 ,

where ζ = ( ζ 1 , ζ 2 ) T 2 , δ ̃ = ( δ 1 , δ 2 ) ( 0,2 ) 2 . It is well understood that the Volume of the Carleson box in two dimensions behaves like V a ( S ( ζ , δ ) ) δ 1 a + 2 δ 2 a + 2 . The above lemma stated in [12] was the main tool in the papers of Kosiński [2], Bayart [3],4],11],13]. The main results regarding composition operators induced by RIFs are the following.

Theorem 2.2.

Let Φ = (ϕ, ψ) with ϕ = p 1 ̃ p 1 and ψ = p 2 ̃ p 2 . Assume that both polynomials p 1, p 2 have one zero on T 2 . Then there exists a q > 0, such that the composition operator C Φ : A β 2 q 2 2 ( D 2 ) A β 2 ( D 2 ) is bounded for all β > 2q.

The second Theorem is.

Theorem 2.3.

Let Φ O ( D 2 , D 2 ) , Φ = (φ, φ) where φ is a RIF. Assume that φ ( z 1 , z 2 ) = p ̃ ( z 1 , z 2 ) p ( z 1 , z 2 ) where p C [ z 1 , z 2 ] is a polynomial which is stable on D 2 , with a single zero τ T 2 . Then, C Φ : A 2 ( D 2 ) A 2 ( D 2 ) is not bounded.

Note here that non-boundedness here is somehow the expected result, in the sense that the self map collapses on the diagonal of D 2 and this makes the volume of the inverse image of the Carleson box under Φ behave similarly to the volume of a one dimensional disc.

3 An example

In the following example we estimate the volume of the sub-level sets { z D 2 : | ϕ ( z ) 1 | < δ } for ϕ(z) being the Knese function, namely ϕ ( z ) = 2 z 1 z 2 z 1 z 2 2 z 1 z 2 , z 1 , z 2 D .

Example 3.1.

Let Φ ( z 1 , z 2 ) = 2 z 1 z 2 z 1 z 2 2 z 1 z 2 , 2 z 1 z 2 z 1 z 2 2 z 1 z 2 . Then C Φ : A β 2 ( D 2 ) A β 2 ( D 2 ) is not bounded, for all β ≥ −1.

Proof.

It suffices to find only one Carleson box for which the condition V β ( Φ 1 ( S ( ζ , δ ̃ ) ) ) C V β ( S ( ζ , δ ̃ ) ) fails, for all constants C > 0 and for all δ ̃ ( 0,1 ) 2 . Consider ζ = ( 1,1 ) T 2 and δ ̃ = ( δ 1 , δ 1 ) . Then,

V β ( Φ 1 ( S ( ( 1,1 ) , δ ̃ ) ) ) = V β z D 2 : | ϕ ( z ) 1 | < δ 1 .

We will estimate the volume on the right hand side. This is equivalent to calculate the volume

V β z D 2 : 2 ( 1 z 1 ) ( 1 z 2 ) 2 z 1 z 2 < δ 1 .

At this moment, set z 1 = 1 + u and z 2 = 1 + v. Then, we see that | ϕ ( z ) 1 | = 2 u v u + v . This is equivalent to estimate the volume of the set

V β { u , v : | u v | < δ 1 | u + v | } .

Consider now points u, v such that |v|≥ C > 0 and | u | δ 1 4 , restricted to D(0, r) × D(0, r) where r > C > 0. For such points, if |u| ≤ |v|/2, then |u + v|≥|v| − |u|≥|v|/2. Since |u| < δ 1/4 and δ 1 is small, |u| ≤ |v|/2 holds. Then | u v | < δ 1 4 | v | = δ 1 2 | v | 2 δ 1 2 | u + v | . Consequently, the set A δ 1 = { ( u , v ) D ( 0 , r ) × D ( 0 , r ) : | v | C , | u | < δ 1 / 4 } is contained in the preimage under Φ of the Carleson box S ( ( 1,1 ) , δ ̃ ) . Hence,

(3.1) V β ( Φ 1 ( S ( ( 1,1 ) , δ ̃ ) ) ) > V β ( A δ 1 ) = | v | C d A β ( v ) | u | < δ / 4 d A β ( u ) = π ( r 2 C 2 ) 16 δ 1 2 + β .

This showcases that such a composition operator is not bounded. □


Corresponding author: Athanasios Beslikas, Doctoral School of Exact and Natural Studies, Institute of Mathematics, Faculty of Mathematics and Computer Science, Jagiellonian University, Łojasiewicza 6, PL30348, Cracow, Poland, E-mail: 

Acknowledgements

I would like to thank all the participants in the Bench Math Session 2023 and 2025 for their support in our cause. Also, I would like to thank Carlo Bellavita for prompting me to make the idea of the problem list official, and the editors William T. Ross and Javad Mashreghi for their willingness to support the Problem List. Last but not least, special thanks go to prof. Alan Sola for supporting our workshop and Łukasz Kosiński for his overall help on the organizing obligations of the workshop.

  1. Funding information: The author is financially supported by the National Science Center, Poland, SHENG III, research project 2023/48/Q/ST1/00048.

  2. Author contributions: The author confirms the sole responsibility for the conception of the study, presented results, and manuscript preparation.

  3. Conflict of interest: The author states no conflict of interest.

References

[1] W. Rudin, Function Theory in Polydisks, W. A. Benjamin, Inc., New York Amsterdam, 1969.Search in Google Scholar

[2] Ł. Kosiński, Composition operators on the polydisc, J. Funct. Anal. 284 (2023), no. 5, 109801, https://doi.org/10.1016/j.jfa.2022.109801.Search in Google Scholar

[3] F. Bayart, Composition operators on the polydisk induced by affine maps, J. Funct. Anal. 260 (2011), no. 7, 1969–2003, https://doi.org/10.1016/j.jfa.2010.12.019.Search in Google Scholar

[4] A. Beslikas, Composition operators and rational inner functions on the bidisc, Proc. Am. Math. Soc. 153 (2025), no. 8, 3491–3502, https://doi.org/10.1090/proc/17294.Search in Google Scholar

[5] K. Bickel, J. A. Cima, and A. Sola, Clark measures for rational inner functions, Michigan Math. J. 73 (2023), no. 5, 1021–1057, https://doi.org/10.1307/mmj/20216046.Search in Google Scholar

[6] K. Bickel and G. Knese, Inner functions on the bidisk and associated hilbert spaces, J. Funct. Anal. 265 (2013), no. 11, 2753–2790, https://doi.org/10.1016/j.jfa.2013.08.002.Search in Google Scholar

[7] K. Bickel, J. E. Pascoe, and A. Sola, Singularities of rational functions in higher dimensions, Am. J. Math. JHU Press 144 (2022), no. 4.10.1353/ajm.2022.0025Search in Google Scholar

[8] K. Bickel, J. E. Pascoe, and A. Sola, Derivatives of rational inner functions and integrability at the boundary, Proc. Lond. Math. Soc. 116 (2015), no. 2, 281–329.10.1112/plms.12072Search in Google Scholar

[9] G. Knese, Rational inner functions in the schur-agler class of the polydisc, Publicacions Matemàtiques 55 (2011), no. 2, 343–357, https://doi.org/10.5565/publmat_55211_04.Search in Google Scholar

[10] G. Knese, Integrability and regularity of rational inner functions, Proc. Lond. Math. Soc. 111 (2015), no. 6, 1261–1306, https://doi.org/10.1112/plms/pdv061.Search in Google Scholar

[11] A. Beslikas, Composition operators and rational inner functions II: boundedness between two different bergman spaces, https://doi.org/10.48550/arXiv.2509.04366.Search in Google Scholar

[12] H. Koo, M. Stessin, and K. Zhu, Composition operators on the polydisc induced by smooth symbols, J. Funct. Anal. 254 (2008), 2911–2925, https://doi.org/10.1016/j.jfa.2008.03.003.Search in Google Scholar

[13] F. Bayart, Composition operators on the hardy space of the tridisc, Ind. Univ. Math. J., (to appear).Search in Google Scholar

Received: 2025-11-06
Accepted: 2026-02-11
Published Online: 2026-03-20

© 2026 the author(s), published by De Gruyter, Berlin/Boston

This work is licensed under the Creative Commons Attribution 4.0 International License.

Downloaded on 1.5.2026 from https://www.degruyterbrill.com/document/doi/10.1515/conop-2025-0018/html?lang=en
Scroll to top button