Field choice problem in persistent homology

DOI10.1007/S00454-023-00544-7MaRDI QIDQ6050218zbMATH OpenOpenAlexFDO

Authors Ippei Obayashi, Michio Yoshiwaki

Publication date 12 October 2023

Published in Discrete \& Computational Geometry (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/1911.11350

zbMATH Keywords

algebraic topologyalgorithmpersistent homologytopological data analysis

Mathematics Subject Classification ID

Persistent homology and applications, topological data analysis (55N31) Applied homological algebra and category theory in algebraic topology (55U99)

Abstract: This paper tackles the problem of coefficient field choice in persistent homology. When we compute a persistence diagram, we need to select a coefficient field before computation. We should understand the dependency of the diagram on the coefficient field to facilitate computation and interpretation of the diagram. We clarify that the dependency is strongly related to the torsion of $mathbbZ$ relative homology in the filtration. We show the sufficient and necessary conditions of the independence of coefficient field choice. An efficient algorithm is proposed to verify the independence. In a numerical experiment with the algorithm, a persistence diagram rarely changes even when the coefficient field changes if we consider a filtration in $mathbb{R}^3$. The experiment suggests that, in practical terms, changes in the field coefficient will not change persistence diagrams when the data are in $mathbb{R}^3$.

Review text

The choice of coefficients in homology is an important consideration when analysing data sets using persistent homology. Often, \(\mathbb{Z}_{2}\) coefficients are used as they have algorithmic advantages. It is not clear what effects this choice of coefficients has on the analysis of data sets. In particular, under which circumstances the persistence diagrams are independent of the choice of coefficients. In this paper, the authors pose 5 questions: 1) are there conditions such that the persistence diagrams are independent of the coefficient field? 2) if such conditions exist, can they be checked algorithmically? 3) should the persistence diagrams for a data set be dependent on the choice of coefficients, how can the diagrams be computed for multiple coefficient fields? 4) with what frequency does the diagram change as we change coefficients? 5) when the persistence diagram does change, how does it change? Questions 1 and 2 are answered completely, while Questions 4 and 5 are partially answered by the authors. They provide an algorithm to check the dependence of the persistence diagrams on the choice of coefficient field.

Recommendations

• Computing persistent homology
• Combinatorial presentation of multidimensional persistent homology
• Computing persistent homology

Cites work

• \textsc{Phat} -- persistent homology algorithms toolbox
• Cohen-Lenstra heuristics for torsion in homology of random complexes
• Computational topology. An introduction
• Computing persistent homology
• Homological algebra for persistence modules
• Homological connectivity of random 2-complexes
• scientific article; zbMATH DE number 5506200 (Why is no real title available?)
• scientific article; zbMATH DE number 1052006 (Why is no real title available?)
• scientific article; zbMATH DE number 2103273 (Why is no real title available?)
• scientific article; zbMATH DE number 3235051 (Why is no real title available?)
• Minimum spanning acycle and lifetime of persistent homology in the Linial-Meshulam process
• On the local behavior of spaces of natural images
• Persistence modules with operators in Morse and Floer theory
• Proximity of persistence modules and their diagrams
• Sliding windows and persistence: an application of topological methods to signal analysis
• Smooth surfaces for multi-scale shape representation
• Stability of persistence diagrams
• Three-dimensional alpha shapes
• Topological persistence and simplification
• Topology and data
• Topology of viral evolution
• Zigzag persistent homology in matrix multiplication time

Cited in

(4)

• Relative persistent homology
• Exact computation of the matching distance on 2-parameter persistence modules
• Interval decomposition of infinite persistence modules over a principal ideal domain and field choice in persistent homology
• Interval decomposition of persistence modules over a principal ideal domain

This page was built for publication: Field choice problem in persistent homology

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6050218)