Correspondences on Riemann Surfaces and Non-Uniform Hyperbolicity

Basic Information

• Paper ID: 2407.15548
• Title: Correspondences on Riemann surfaces: (non-uniform) hyperbolicity and graph attractors
• Authors: Laurent Bartholdi, Dzmitry Dudko, Kevin M. Pilgrim
• Classification: math.DS (Dynamical Systems)
• Publication Date: July 2024, last revised October 2025
• Paper Link: https://arxiv.org/abs/2407.15548

Abstract

This paper investigates certain correspondences on Riemann surfaces, proving they possess a form of weak hyperbolicity: sufficiently long loops at fixed points become shorter when lifted and closed. Within the algebraic encoding framework of bisets, this translates to contraction properties of fundamental group elements along iterative lifting sequences. As an application, the authors prove that, except for the usual Lattès counterexamples, there exists a finite invariant set of curve homotopy classes for rational maps with 4 postcritical points, with each curve attracted to this set under iterative lifting. More generally, for graphs of given complexity, there exists a finite invariant set of graph homotopy classes to which every graph is attracted.

Research Background and Motivation

Problem Background

The core problem investigated in this paper originates from a fundamental question in Thurston theory: understanding the combinatorial structure of post-critically finite (PCF) rational maps. Specifically:

1. Generalization from surface homeomorphisms to branched coverings: Many key aspects of surface homeomorphism theory have been generalized to post-critically finite branched self-coverings on $S^2$ (i.e., Thurston maps). This includes geometric obstructions and actions on Teichmüller spaces.
2. Specificity of the polynomial case: The case of PCF hyperbolic polynomials has been well understood, with forward-invariant Hubbard trees serving as complete combinatorial invariants. Under lifting operations, all curves converge exponentially fast to finite invariant sets.
3. Difficulties in the general case: For general PCF rational maps, analysis becomes extremely complex due to the existence of obstructions to twisting $g_0 \circ f \circ g_1$, which prevent the mapping class biset from possessing contractivity in the Nekrashevych sense.

Research Motivation

The authors' primary motivation is to develop a similarly satisfactory theory beyond the polynomial case. The central difficulty lies in analyzing the tension between two opposing forces:

• "Non-uniform contraction": induced by lifting operations
• "Additive corrections": ensuring the analyzed process is iterable

Core Contributions

1. Main theoretical result (Theorem A): Proves the existence of finite curve attractors for non-Lattès rational maps with 4 postcritical points, where non-uniform contraction eventually dominates.
2. Technical core (Theorem C): Establishes crucial contraction estimates for admissible correspondences, providing the technical foundation for the main results.
3. Application to graph theory: Extends results to finite attractors for multicurves, trees, and spine graphs.
4. Algebraic framework: Provides a unified algebraic treatment through biset theory.
5. Strategy for higher dimensions: Proposes concrete research strategies for handling cases with more than 4 postcritical points.

Methodology Details

Core Concept Definitions

Admissible complex correspondences: Correspondences $\phi, \rho: T \Rightarrow S$ satisfying:

• $T, S$ are hyperbolic Riemann surfaces
• $\phi: T \to S$ is an analytic covering map of finite degree
• $\rho: T \to S$ is an analytic map but not a covering map
• $\rho$ is a contraction in the hyperbolic metric

X-rays: Given a basepoint $\star \in S$ and a finite set of paths $X$, an X-ray is an infinite sequence $g^{(0)}, g^{(1)}, g^{(2)}, \ldots$ satisfying: $g^{(n)} \leadsto \rho \circ \phi_*(g^{(n)}) \cdot \overline{x_n} =: g^{(n+1)}$

Main Technical Framework

Thick-thin decomposition theory: For a fundamental group element $g \in \pi_1(S, \star)$, there exists a decomposition: $g = \ell_0 \cdot r_1 \cdot \ell_1 \cdots r_k \cdot \ell_k$ where:

• $r_i$ are "roundabouts" located in neighborhoods of cusps
• $\ell_i$ are "thick segments" located in the thick part

Key technical estimates: For a roundabout $r_i$, the length formula is: $|r_i| = 2\log^+ |\angle(r_i)| + O(1)$ where $\angle(r_i)$ is the winding number of $r_i$ around the cusp.

Proof Strategy

Proof by contradiction framework:

1. Assume Theorem C fails, constructing "tight sequences" $(g_m)$ satisfying $|g_m| > m$ and $|g_m^{(n)}| > |g_m| - 1/m$
2. Extraction argument: Extract "very tight" subsequences from tight sequences, maximizing the number of unbounded roundabouts to $K$
3. Local degree analysis: For consecutive roundabouts $r_{i,m}, r_{i+1,m}$, prove that local degrees satisfy: $t_i^{(n)} t_{i+1}^{(n)} = O(\exp(-\eta n))$
4. Derivation of contradiction: Combining with the length formula: $|g_m^{(n)}| \leq |g_m| - 2\eta n + O(1)$ contradicting the definition of tight sequences.

Experimental Setup

Specific Examples

Rabbit polynomial: $f(z) = z^2 + c$ where the critical point $z=0$ has period 3. The corresponding moduli space correspondence is $F(x) = 1 - 1/x^2$, which is uniformly hyperbolic.

Tree-like polynomial: $f(z) = z^2 + i$ with correspondence $F(x) = (-1 + 2/x)^2$, whose Julia set is the entire sphere.

Lodge example: $f(z) = 3z^2/(2z^3 + 1)$, providing an example of non-degree-hyperbolic correspondence.

Critically fixed maps: Quintic and cubic critically fixed maps with local degrees $(2,3,3,4)$ and $(2,2,2,2)$.

Evaluation Metrics

• Contraction constants: Measuring the degree of contraction of lifting operations
• Attractor size: Number of elements in finite attractors
• Complexity bounds: Upper bounds on the sum of vertices and edges in graphs

Experimental Results

Main Theorems

Theorem A (Finite attractors for X-rays): Let $F = \phi, \rho: T \Rightarrow S$ be an admissible correspondence and $\star \in S$ a fixed point of $F$. For each finite set $X$, there exists a finite attractor $A(X) \subset \pi_1(S, \star)$ such that every X-ray $g^{(0)}, g^{(1)}, g^{(2)}, \ldots$ eventually enters $A(X)$.

Corollary B (Case of 4 postcritical points): Let $\#P = 4$ and $f$ be a non-Lattès rational map. Then finite attractors exist for pullback iterations of multicurves, trees, and spines.

Technical Core Results

Theorem C (Additive contraction estimates): There exist constants $\kappa > 0, N \in \mathbb{N}, \epsilon > 0, \xi > 0$ such that:

1. $|g^{(n+1)}| \leq |g^{(n)}| + \xi$
2. If $|g| > \kappa$ then there exists $n \in \{0, \ldots, N\}$ such that $|g^{(n)}| \leq |g| - \epsilon$

Related Work

Historical Development

• Thurston theory: Classification theory of post-critically finite branched coverings
• Hyperbolic dynamics: Theory of uniformly hyperbolic systems
• Teichmüller theory: Dynamics on moduli spaces

Related Results

• Nekrashevych's combinatorial model theory
• Belk et al.'s work on topological polynomial recognition
• Cui-Gao-Zeng's work on invariant graphs

Conclusions and Discussion

Main Conclusions

1. Demonstrates the effectiveness of non-uniform hyperbolicity in the theory of complex correspondences
2. Establishes existence of finite attractors in the case of 4 postcritical points
3. Provides concrete strategies for handling more general cases

Limitations

1. Current results are limited to the case of 4 postcritical points
2. Lattès maps require special treatment
3. Higher-dimensional generalizations remain conjectural

Future Directions

1. Conjecture D: Generalize results to arbitrary numbers of postcritical points
2. Develop higher-dimensional decomposition theory
3. Understand interactions between parameter space and dynamical space

In-Depth Evaluation

Strengths

1. Theoretical depth: Cleverly combines hyperbolic geometry, complex dynamics, and group theory
2. Technical innovation: The methods of thick-thin decomposition and extraction arguments are general
3. Broad applicability: Results apply to multicurves, trees, graphs, and other geometric objects
4. Clear exposition: Well-structured paper with detailed technical exposition

Weaknesses

1. Limited scope: Main results apply only to the case of 4 postcritical points
2. Technical complexity: Proofs are highly technical and difficult to generalize
3. Computational complexity: Computing attractors in practice may be difficult

Impact

1. Theoretical contribution: Provides new perspectives on combinatorial problems in complex dynamics
2. Methodological value: The treatment of non-uniform hyperbolicity has universal significance
3. Future development: Lays foundation for solving more general cases

Applicable Scenarios

• Classification problems for post-critically finite rational maps
• Dynamics on moduli spaces
• Study of contraction properties in geometric group theory

References

The paper cites 42 important references spanning complex dynamics, Thurston theory, hyperbolic geometry, and other fields, reflecting the interdisciplinary nature and theoretical depth of this research.