Mixing and ergodicity of compositions of inner functions

Basic Information

• Paper ID: 2405.06411
• Title: Mixing and ergodicity of compositions of inner functions
• Authors: Gustavo R. Ferreira, Artur Nicolau
• Classification: math.DS (Dynamical Systems), math.CV (Complex Variables)
• Publication Date: May 10, 2024 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2405.06411

Abstract

This paper investigates the ergodic and mixing properties of non-autonomous dynamical systems on the unit circle generated by sequences of inner functions fixing the origin.

Research Background and Motivation

Research Problem

The core problem addressed in this paper is the analysis of ergodicity and mixing properties of non-autonomous dynamical systems $G_n = g_n \circ g_{n-1} \circ \cdots \circ g_1$ on the boundary of the unit disk, where each $g_n$ is an inner function fixing the origin.

Problem Significance

1. Theoretical Importance: Non-autonomous inner function dynamics naturally arise in complex dynamics, particularly in the study of simply connected wandering domains of entire functions
2. Extension of Classical Theory: Traditional ergodic theory primarily focuses on autonomous systems ($T_n = T^n$), while research on non-autonomous systems remains relatively limited
3. Measure-Theoretic Dynamics: Boundary mappings of inner functions preserve Lebesgue measure, providing a favorable measure-theoretic framework for investigation

Limitations of Existing Methods

1. Constraints of Autonomous Systems: Classical Denjoy-Wolff theorem and related ergodic theory primarily address iterations of single inner functions
2. Complexity of Non-Autonomous Systems: In the non-autonomous setting, even when the system is mixing in the usual sense, time averages may not converge to spatial averages
3. Lack of Unified Theory: Systematic characterization of ergodic properties of inner function compositions is lacking

Research Motivation

The authors adopt the definitions of Berend and Bergelson to study ergodicity and mixing of non-autonomous systems, which preserve the classical properties that "mixing implies ergodicity" and "mixing is equivalent to ergodicity of arbitrary subsequences."

Core Contributions

1. Complete Characterization of Ergodicity: Provides necessary and sufficient conditions for ergodicity of non-autonomous inner function composition sequences (Theorem 1.2)
2. Discrimination Criteria for Specific Cases: Establishes explicit ergodicity and mixing conditions for cases where derivatives are positive real numbers and non-contracting sequences (Theorem 1.3)
3. Converse of Pommerenke's Result: Establishes equivalence between contractivity and mixing in the usual sense (Theorem 5.2)
4. Construction of Counterexamples: Provides instances of contracting but non-ergodic sequences, illustrating distinctions between different mixing concepts
5. Recurrence Theory: Proves general results that ergodicity implies recurrence (Theorem 1.6)

Methodology Details

Problem Formulation

Let $g_n: \mathbb{D} \to \mathbb{D}$ be a sequence of inner functions fixing the origin. Define the composition functions: $G_n = g_n \circ g_{n-1} \circ \cdots \circ g_1, \quad G_n^m = g_n \circ \cdots \circ g_{m+1}$

The study focuses on ergodicity and mixing properties of the boundary mapping sequence $\{\hat{G}_n\}_{n \in \mathbb{N}}$ on the unit circle $\partial\mathbb{D}$.

Core Definitions

Definition 1.1 (Ergodicity and Mixing for Non-Autonomous Systems):

• A sequence $\{T_n\}_{n \in \mathbb{N}}$ is ergodic if for any $\varphi \in L^2(\mu)$: $\lim_{N \to \infty} \left\|\frac{1}{N}\sum_{n=1}^N \varphi \circ T_n - \int_X \varphi d\mu\right\|_2 = 0$
• A sequence $\{T_n\}_{n \in \mathbb{N}}$ is mixing if every subsequence is ergodic.

Main Technical Tools

1. Relationship Between Fourier Coefficients and Derivatives

Lemma 3.2: For compositions of inner functions, $\langle e_\ell \circ \hat{G}_n, e_\ell \circ \hat{G}_m \rangle = ((G_n^m)'(0))^\ell$

This key equation connects the $L^2$ inner product with the derivative of the composition function at the origin.

2. Characterization of Ergodicity

Theorem 1.2 (Necessary and Sufficient Conditions for Ergodicity): The sequence $\{\hat{G}_n\}_{n \in \mathbb{N}}$ is ergodic if and only if for any $\ell \in \mathbb{N}$: $\lim_{N \to \infty} \Re\left(\frac{1}{N^2}\sum_{m=1}^{N-1}\sum_{n=m+1}^N ((G_n^m)'(0))^\ell\right) = 0$

Technical Innovations

1. Derivative Characterization Method: Characterizes ergodicity through properties of composition function derivatives, avoiding direct treatment of measure-theoretic complexity
2. Case-by-Case Analysis: Provides specialized discrimination criteria for different derivative properties (positive real numbers, non-contracting, etc.)
3. Application of Equidistribution Theory: In the non-contracting case, connects ergodicity with equidistribution of the argument of derivatives

Experimental Setup

Theoretical Verification

This paper is primarily theoretical research, with verification through construction of specific examples:

1. Example of Contracting but Non-Ergodic Sequence (Proposition 6.1):
   • Constructs Blaschke products $g_n(z) = z \cdot \frac{z + a_n}{1 + a_n z}$, where $a_n = \frac{n}{n+1}$
   • Verifies $\sum_{n \geq 1}(1-|g_n'(0)|) = \infty$ (contractivity)
   • But does not satisfy ergodicity conditions
2. Specific Ergodic and Mixing Examples (Corollary 6.2):
   • Provides concrete conditions ensuring mixing
   • Analyzes ergodicity in random settings

Experimental Results

Main Results

1. Case of Positive Real Derivatives (Theorem 1.3(i)(ii))

When all $g_n'(0) > 0$:

• Ergodicity: Equivalent to for any $0 < \varepsilon < 1$, $\lim_{N \to \infty} \prod_{k=\lfloor N(1-\varepsilon)\rfloor}^N g_k'(0) = 0$
• Mixing: Equivalent to for any $\epsilon > 0$, there exists $M_0$ such that for $N > M > M_0$, $\prod_{k=N-M}^N g_k'(0) < \epsilon$

2. Non-Contracting Case (Theorem 1.3(iii))

When the sequence is non-contracting:

• Ergodicity is equivalent to the sequence $\{e^{i\arg G_n'(0)}\}_{n \in \mathbb{N}}$ being equidistributed on $\partial\mathbb{D}$

3. Relationship Between Contractivity and Mixing

Theorem 5.2: The sequence $\{\hat{G}_n\}_{n \in \mathbb{N}}$ is mixing in the usual sense if and only if $\{g_n\}_{n \in \mathbb{N}}$ is contracting.

Corollary 1.5: $\{g_n\}_{n \in \mathbb{N}}$ is contracting if and only if $\{\hat{G}_n\}_{n \in \mathbb{N}}$ has mixing subsequences.

Counterexample Analysis

The example constructed in Proposition 6.1 demonstrates:

• Sequences can be contracting and mixing in the usual sense
• But may not be ergodic under the strong definition
• This illustrates essential distinctions between different mixing concepts

Related Work

Classical Theoretical Foundations

1. Denjoy-Wolff Theorem: Provides foundations for convergence of inner functions within the unit disk
2. Loewner Lemma: Ensures that boundary mappings of inner functions preserve Lebesgue measure
3. Classical Ergodic Theory: Foundational work by Aaronson, Pommerenke, Doering-Mañé, and others

Non-Autonomous Dynamics

1. Berend-Bergelson Theory: Provides appropriate definitions of ergodicity and mixing for non-autonomous systems
2. Applications in Complex Dynamics: Non-autonomous inner function dynamics naturally arise in the study of wandering domains

Innovations of This Paper

Compared to existing work, this paper:

1. Systematically studies ergodic properties of non-autonomous inner function compositions for the first time
2. Establishes a complete theoretical framework
3. Provides computable discrimination criteria

Conclusions and Discussion

Main Conclusions

1. Dichotomy Phenomenon: The mechanisms of ergodicity primarily consist of two types—"expanding maps" and "irrational rotations"
2. Quantified Contractivity: Conditions (3) and (4) can be understood as quantified versions of contractivity
3. Recurrence: Ergodicity guarantees recurrence and density of orbits

Limitations

1. Computational Complexity: Discrimination conditions involve infinite products and series, making practical computation potentially difficult
2. General Inner Functions: Theory primarily addresses inner functions fixing the origin; more general cases remain to be studied
3. Higher-Dimensional Generalization: Results are limited to one-dimensional settings

Future Directions

1. Study compositions of inner functions not fixing the origin
2. Explore higher-dimensional analogues
3. Develop more computationally tractable discrimination criteria
4. Investigate relationships with other dynamical properties

In-Depth Evaluation

Strengths

1. Theoretical Completeness: Establishes a complete theoretical system from necessary and sufficient conditions to specific discrimination criteria
2. Technical Innovation: Cleverly employs Fourier analysis and complex analysis tools to address ergodic theory problems
3. Profound Results: Reveals the essential mechanisms of ergodicity in non-autonomous systems
4. Counterexample Construction: Clarifies subtle distinctions between different concepts through concrete examples

Weaknesses

1. Limited Applications: Practical application scenarios for theoretical results are relatively restricted
2. Computational Complexity: Discrimination conditions may be difficult to verify in practice
3. Generalization Challenges: Extension of results to more general settings may face technical difficulties

Impact

1. Theoretical Contribution: Provides important tools for non-autonomous dynamical systems theory
2. Interdisciplinary Connections: Links complex analysis, ergodic theory, and dynamical systems
3. Foundation for Future Research: Establishes groundwork for further investigation of related problems

Applicable Scenarios

1. Complex Dynamics: Study of dynamical properties of wandering domains of entire functions
2. Ergodic Theory: Theoretical analysis of non-autonomous systems
3. Harmonic Analysis: Measure-theoretic problems related to inner functions

References

The paper cites important literature in ergodic theory, complex dynamics, and inner function theory, including:

• Berend & Bergelson (1984): Ergodic theory of non-autonomous systems
• Pommerenke (1981): Ergodic properties of inner functions
• Doering & Mañé (1991): Dynamics of inner functions
• Recent related work such as BEF+22, Fer23, and others

This paper is theoretically significant, providing a systematic ergodic theory framework for non-autonomous inner function dynamics. While applications are relatively specialized, it makes substantial contributions to the development of related fields.