On the automorphism group of the Morse complex

Maxwell Lin, Nicholas A. ScovilleAdv. Appl. Math 2021 · OpenReview

TL;DR

This paper investigates the automorphism group of the Morse complex, a structure derived from discrete Morse theory. The authors establish a precise relationship between the automorphism group of a simplicial complex and its corresponding Morse complex, showing they are often isomorphic, with specific exceptions for cycles and boundaries of simplices. This work provides a complete understanding of the symmetries of these Morse complexes.

What Problem Does This Paper Address?

Discrete Morse theory provides a powerful tool for studying the topology of simplicial complexes by associating them with gradient vector fields. The Morse complex, introduced by Chari and Joswig, is a construction based on these gradient vector fields. While the structure of the Morse complex itself has been studied, its automorphism group – the set of structure-preserving transformations – has remained largely unexplored. Understanding these symmetries is crucial for a deeper comprehension of the Morse complex and its relationship to the original simplicial complex. This paper aims to fill this gap by characterizing the automorphism group of the Morse complex for various classes of simplicial complexes.

Methods

The paper's core methodology involves establishing connections between automorphisms of a simplicial complex $K$ and automorphisms of its Morse complex $M(K)$.

  • Induced Automorphisms: The authors define an "induced automorphism" on $M(K)$ that arises from a simplicial automorphism of $K$. They prove that this induced map is indeed a simplicial automorphism of $M(K)$ and that the map from $Aut(K)$ to $Aut(M(K))$ is an injective homomorphism.
  • Isomorphism via Hasse Diagram: A key technical tool is the Hasse diagram of a simplicial complex, which is a partially ordered set of its simplices. The paper establishes an isomorphism between the automorphism group of the Morse complex $M(K)$ and the automorphism group of its Hasse diagram $H(K)$. This isomorphism is crucial for simplifying the analysis of $Aut(M(K))$.
  • Case Analysis: The problem is tackled by considering different types of simplicial complexes:
  • General Case: For complexes that are neither cycles ($C_n$) nor boundaries of simplices ($\partial \Delta^n$), the authors show that $Aut(M(K)) \cong Aut(K)$.
  • Cycles ($C_n$): For cycles, they prove $Aut(M(C_n)) \cong Aut(C_{2n})$.
  • Boundaries of Simplices ($\partial \Delta^n$): For boundaries of simplices, they show $Aut(M(\partial \Delta^n)) \cong Aut(\partial \Delta^n) \times \mathbb{Z}_2$.
  • Ghost Automorphisms: For the case of $\partial \Delta^n$, they introduce "ghost automorphisms" which are not induced by simplicial maps on the original complex. These are related to a "reflection map" on $\partial \Delta^n$.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • Automorphism Group Connection: The most significant insight is the precise characterization of $Aut(M(K))$ in terms of $Aut(K)$ for various classes of simplicial complexes. This provides a complete picture of the symmetries of Morse complexes.
  • Hasse Diagram as a Bridge: The isomorphism $Aut(M(K)) \cong Aut(H(K))$ is a powerful technical tool. It allows the authors to leverage the structure of the Hasse diagram to understand the automorphisms of the Morse complex, simplifying complex combinatorial problems.
  • "Ghost" Automorphisms: The discovery of "ghost" automorphisms for $\partial \Delta^n$ is a key finding. These are automorphisms of the Morse complex that do not arise from simple symmetries of the original complex, revealing a richer structure in $M(\partial \Delta^n)$.
  • Reflection Map: The introduction and analysis of the reflection map on $\partial \Delta^n$ and its induced action on the Morse complex are central to understanding the $\mathbb{Z}_2$ factor in $Aut(M(\partial \Delta^n))$.

Key Results

The paper's main result is summarized in Theorem 1:

  • For a finite, connected, abstract simplicial complex $K$:
  • If $K$ is neither a cycle ($C_n$) nor the boundary of an $n$-simplex ($\partial \Delta^n$), then $Aut(M(K)) \cong Aut(K)$.
  • If $K = C_n$, then $Aut(M(C_n)) \cong Aut(C_{2n})$.
  • If $K = \partial \Delta^n$, then $Aut(M(\partial \Delta^n)) \cong Aut(\partial \Delta^n) \times \mathbb{Z}_2$.

Limitations and Future Work

  • Finite Complexes: The results are stated for finite simplicial complexes. Extending these to infinite complexes might be a direction for future work.
  • Connected Complexes: The paper assumes connected simplicial complexes. The behavior for disconnected complexes is not explicitly addressed.
  • Generalization of Ghost Automorphisms: While ghost automorphisms are identified for $\partial \Delta^n$, their nature and prevalence in other complex structures could be a topic for further research.
  • Computational Aspects: The paper focuses on theoretical characterization. Developing efficient algorithms to compute these automorphism groups for specific, complex inputs could be a practical extension.

Summary updated

BibTeX
@inproceedings{lin2021automorphism,
  author = {Maxwell Lin and Nicholas A. Scoville},
  title = {{On the automorphism group of the Morse complex}},
  booktitle = {Adv. Appl. Math},
  year = {2021},
  url = {https://openreview.net/forum?id=0D9doFrTSp9},
}

Loading related content...

Figure