- The paper proves that the composition operator is compact if and only if an integral condition involving the Nevanlinna counting function and the inner function’s defect vanishes as the boundary is approached.
- By applying real interpolation theory, the work demonstrates that compactness in the Hilbert space case (p=2) extends to all p in (1,∞), confirming the p-independence of the results.
- The study further characterizes compactness for one-component inner functions through Clark measures, linking the vanishing singular parts of these measures to operator compactness.
Compactness of Composition Operators Between Model and Hardy Spaces
Introduction
This paper, "Composition operators between model and Hardy spaces" (2604.04054), systematically investigates the structure and compactness criteria of composition operators acting between model spaces KΘp and Hardy spaces Hp(Dn). For a holomorphic symbol φ:Dn→D and an inner function Θ on the unit disk, the core question addressed is: for which φ is the composition operator Cφ:KΘp→Hp(Dn) compact? The analysis employs advanced tools—Nevanlinna counting functions and Clark measures—achieving complete, technically nuanced characterizations in the general polydisk context.
Preliminaries and Notation
Hardy spaces Hp(Dn) and model spaces KΘp are central to the discourse. Recall,
- Hp(Dn) denotes the space of holomorphic functions on the polydisk Dn that satisfy the standard Hp(Dn)0 growth condition on boundary approach.
- Given an inner function Hp(Dn)1 on Hp(Dn)2, Hp(Dn)3, with Hp(Dn)4 being the classical model space Hp(Dn)5.
- The composition operator Hp(Dn)6 is defined by Hp(Dn)7.
The compactness of Hp(Dn)8 is tied intricately to the analytic and geometric properties of the symbol Hp(Dn)9 and the inner function φ:Dn→D0.
Main Compactness Criteria
Nevanlinna Counting Function Approach
A central role is played by the Nevanlinna counting function φ:Dn→D1, which encodes the multiplicity of preimages of φ:Dn→D2 under φ:Dn→D3 with a logarithmic weight favoring proximity to the boundary. The following equivalence is established:
- φ:Dn→D4 is compact if and only if
φ:Dn→D5
Here φ:Dn→D6 is the normalized measure on the torus φ:Dn→D7 and φ:Dn→D8 approaches the distinguished boundary. This result generalizes classical one-variable theorems and highlights the interplay between the symbolic dynamics of φ:Dn→D9 and the defectiveness of Θ0 near the boundary.
Compactness Independence of Θ1
By leveraging real interpolation theory, specifically the Cwikel compactness theorem for interpolation spaces, the paper demonstrates that compactness of Θ2 is independent of the value of Θ3 in Θ4. In particular, compactness in the Hilbertian case Θ5 implies compactness for all Θ6, and vice versa. This sharp, general result is nontrivial due to the structural complexities of Θ7 for Θ8.
Clark Measure Criterion for One-Component Inner Functions
The paper provides an alternative and equally complete characterization for one-component inner functions using Clark measures. For such Θ9, φ0 is compact if and only if the singular parts of all Clark measures vanish on the spectrum of φ1. That is,
φ2
where φ3 is the spectrum, and φ4 is the Clark measure associated to φ5. This links operator-theoretic compactness to a spectral property of singular measures arising from the inner function, extending and unifying prior results for the disk.
Technical Contributions
- Stanton’s formula is generalized for φ6, allowing explicit expressions of φ7-norms post-composition in terms of the Nevanlinna function.
- Subharmonicity of the Nevanlinna counting function is used to establish necessary integral estimates.
- Detailed kernel estimates for model spaces are utilized in the Clark measure direction, especially when sequences tend to the boundary point spectrum.
- The one-sided compactness result for real interpolation (Cwikel’s theorem) is a backbone for independence-of-φ8 arguments, enabling extension from Hilbert to Banach settings.
Implications and Further Directions
The compactness characterizations elucidated in this paper rigorously unify perspectives from function theory (via inner functions and Clark measures), operator theory (through compactness and interpolation), and several complex variables (for φ9). The explicit criteria enable, in principle, the determination of compactness for large classes of composition operators arising in function-theoretic and applied settings, such as control theory and systems analysis where model spaces arise.
The dependence on the symbol Cφ:KΘp→Hp(Dn)0 solely through the Nevanlinna counting function suggests future investigations into the sharpness of such criteria, possible relaxations for more general domains, and stability under perturbations of Cφ:KΘp→Hp(Dn)1. Extending the Clark measure criteria to Cφ:KΘp→Hp(Dn)2 and non one-component inner functions remains a compelling theoretical challenge.
Conclusion
The paper achieves a comprehensive solution to the compactness problem for composition operators between model and Hardy spaces in the polydisk, unifying Nevanlinna-type and Clark measure approaches. The independence from the parameter Cφ:KΘp→Hp(Dn)3 in the Banach range adds practical scope to operator classification. These results provide deep links among analytic function theory, harmonic analysis, and operator theory, setting a strong foundation for further advances in the spectral analysis of non-self-adjoint operators and functional analysis on complex domains.