Papers
Topics
Authors
Recent
Search
2000 character limit reached

Hyperbolic spaces with geometric and geometrically finite quasi-actions are symmetric

Published 15 Apr 2026 in math.GR and math.GT | (2604.13898v1)

Abstract: We prove that if a proper metric space is quasi-isometric to a finitely generated group and to a space with a horoball over a finitely generated group, then that space is quasi-isometric to a rank-one symmetric space or the real line.

Summary

  • The paper proves that proper metric spaces with geometric and geometrically finite quasi-actions are quasi-isometric only to rank-one symmetric spaces or ℝ.
  • It employs reduction techniques using pointed Gromov–Hausdorff convergence and asymptotic cone analysis to achieve a precise classification.
  • The findings extend lattice classification to quasi-action settings, linking geometric rigidity with boundary structure and group actions.

Symmetry of Hyperbolic Spaces with Geometric and Geometrically Finite Quasi-Actions

Overview

The paper "Hyperbolic spaces with geometric and geometrically finite quasi-actions are symmetric" (2604.13898) presents a rigorous characterization of proper metric spaces that are quasi-isometric both to finitely generated groups and to spaces equipped with horoballs. The primary result demonstrates that these spaces are, up to quasi-isometry, restricted precisely to rank-one symmetric spaces or R\mathbb{R}, revealing deep connections between large-scale geometry and algebraic properties of groups acting on such spaces. The findings extend previous lattice classification theorems to the quasi-isometric category, delineating the rigidity and uniqueness properties of hyperbolic geometry with respect to geometric and geometrically finite actions.

Main Results

The foundational theorem asserts that if a proper metric space XX' is quasi-isometric both to a finitely generated group Γ\Gamma and to a space containing a horoball constructed over a finitely generated group, then XX' is quasi-isometric to a rank-one symmetric space or R\mathbb{R}. Furthermore, Γ\Gamma is virtually a rank-one uniform lattice in the isometry group of the symmetric space, and whenever the horoball space is quasi-isometric to the cusped space of a relatively hyperbolic group pair (H,P)(H,\mathbb{P}), HH is virtually a non-uniform rank-one lattice.

This classification is achieved through a careful analysis of the structural and growth properties induced by horoballs. The result leverages the recent advances in the quasi-isometry theory of solvable groups, particularly those of Dymarz, Fisher, and Xie, as well as a theorem by Caprace, Cornulier, Monod, and Tessera on isometry groups of negatively curved symmetric spaces. The methodology exploits the reduction to nilpotent groups through growth constraints and the convergence of metric structures to solvable Lie groups equipped with Carnot–Carathéodory metrics.

An explicit reduction shows that the only possible underlying group structure in the horoball construction, when quasi-isometric to a finitely generated group, is virtually nilpotent. These spaces admit a solvable Lie group structure N^αR\widehat{N} \rtimes_{\alpha} \mathbb{R}, where N^\widehat{N} is an asymptotic cone of a nilpotent group, and XX'0 acts via metric dilations.

Implications for Boundaries and Group Actions

One salient corollary concerns group boundaries: if a hyperbolic group XX'1 has a Gromov boundary quasisymmetric to the Bowditch boundary of a finitely generated relatively hyperbolic group pair with infinite parabolic subgroups, then this boundary must be a sphere, and XX'2 is virtually a uniform rank-one lattice. Notably, for groups with 2-sphere boundaries, this yields direct classification as virtually Kleinian groups, linking geometric finiteness, quasi-isometric rigidity, and boundary structure.

The paper's boundary analysis highlights the finer distinctions between homeomorphic and quasisymmetric boundaries, e.g., the Sierpinski carpet boundary case. The results confirm conjectures concerning the uniqueness of boundary types within hyperbolic and relatively hyperbolic group theory, showing that non-symmetric boundaries such as the Sierpinski carpet cannot be quasisymmetrically equivalent across these contexts.

Technical Innovations

The reduction techniques rest on pointed Gromov–Hausdorff convergence and asymptotic cone analysis, following Pansu's work, and demonstrate that the horoball warped product converges to a solvable Lie group. The proof constructs quasi-isometries locally and applies a diagonal argument to extend them globally, ensuring uniform quasi-isometry constants and surjectivity. The stability under quasi-isometry of boundaries and the restriction to nilpotent group actions are proven by exploiting the bounded geometry of balls in the horoball.

Application of Dymarz–Fisher–Xie ensures the existence of geometric group actions on solvable groups, and Caprace–Cornulier–Monod–Tessera's structural theorem identifies the isometry group dichotomy, thus excluding any other quasi-isometric types apart from rank-one symmetric spaces and XX'3.

Numerical and Structural Outcomes

The classification is exclusive – no proper metric space quasi-isometric both to a finitely generated group and a nontrivial horoball over an infinite group can be quasi-isometric to anything other than a rank-one symmetric space or XX'4. The proof is constructive and exhaustive within the framework of geometric group theory.

Practical and Theoretical Consequences

These results contribute to the rigidity paradigm in geometric group theory, demarcating boundaries for the coarse geometric equivalence of spaces. The theoretical implications include:

  • Quasi-Isometric Rigidity: Group pairs with geometric and geometrically finite quasi-actions possess boundary and large-scale geometric properties that uniquely determine their quasi-isometric class.
  • Boundary Classification: The connection between quasi-isometry and boundary quasisymmetry foregrounds boundaries as classifiers for group actions, potentially informing algorithmic boundary identification in geometric group theory.
  • Horoball Construction Limits: The reduction to virtually nilpotent structures implies that attempts to construct more "exotic" spaces with horoballs always revert to these rigid types under quasi-isometric equivalence.
  • Lattice Classification: The correspondence between group actions and lattice types is extended from isometric actions to quasi-actions, impacting the study of discrete and continuous symmetries in negatively curved spaces.

In practical terms, these classification results can streamline analyses in computational group theory (especially in boundary detection), automate subgroup structure detection in hyperbolic spaces, and ensure boundary conditions are accurately modeled when simulating these metric spaces.

Future Directions

Given the rigidity of quasi-isometric classes in these contexts, future research may focus on:

  • Expanding classification to relatively hyperbolic groups with more intricate peripheral structures.
  • Investigating the possible extensions of quasi-actions to higher-rank symmetric spaces and their associated boundaries.
  • Exploring algorithmic implementations for detecting quasi-isometric equivalence, leveraging the uniqueness results for geometric models.
  • Further analysis of the asymptotic cone structures and their impact on the geometry of boundaries and coarse embeddings.

Conclusion

The paper establishes strong rigidity results for proper metric spaces with geometric and geometrically finite quasi-actions, showing they are quasi-isometric solely to rank-one symmetric spaces or the real line. The implications span boundary theory, group action classification, and the structure of spaces with horoballs, providing a comprehensive bridge between coarse geometry and algebraic group properties in hyperbolic settings. These results further reinforce the foundational role of symmetric spaces in geometric group theory and delineate the strict landscape for quasi-isometric models in negatively curved geometry.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Collections

Sign up for free to add this paper to one or more collections.