Stable Barcodes for Multi-Parameter Persistent Homology

Fuente: Zenodo
Saved in:
Bibliographic Details
Main Author: SÉRGIO DE ANDRADE, PAULO
Format: Recurso digital
Published: Zenodo 2025
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866901190974898176
author SÉRGIO DE ANDRADE, PAULO
author_facet SÉRGIO DE ANDRADE, PAULO
contents One-parameter persistent homology is a cornerstone of topological data analysis, offering a stable and complete invariant: the persistence barcode. The extension to multi-parameter persistent homology, while natural for complex data, is hindered by the absence of a simple barcode decomposition, a consequence of the underlying algebra's wild representation type. This has obstructed the development of robust, interpretable invariants. This paper confronts this challenge by introducing a framework to compute stable, barcode-like summaries for multi-parameter persistence modules. Our approach avoids direct module decomposition. Instead, we systematically project the multi-parameter module onto a canonical set of lines, generating a collection of one-parameter modules, each with a well-defined, stable barcode. The novelty of our work lies in a method to aggregate these barcodes into a single, comprehensive feature vector and, crucially, in providing a formal stability guarantee for this summary. We establish a stability theorem proving that our invariant is continuous with respect to the interleaving distance on multi-parameter modules. This result ensures robustness to noise and provides a theoretically sound, computationally tractable method for applying multi-parameter topological analysis to real-world data.
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_17681596
institution Zenodo
language
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle Stable Barcodes for Multi-Parameter Persistent Homology
SÉRGIO DE ANDRADE, PAULO
One-parameter persistent homology is a cornerstone of topological data analysis, offering a stable and complete invariant: the persistence barcode. The extension to multi-parameter persistent homology, while natural for complex data, is hindered by the absence of a simple barcode decomposition, a consequence of the underlying algebra's wild representation type. This has obstructed the development of robust, interpretable invariants. This paper confronts this challenge by introducing a framework to compute stable, barcode-like summaries for multi-parameter persistence modules. Our approach avoids direct module decomposition. Instead, we systematically project the multi-parameter module onto a canonical set of lines, generating a collection of one-parameter modules, each with a well-defined, stable barcode. The novelty of our work lies in a method to aggregate these barcodes into a single, comprehensive feature vector and, crucially, in providing a formal stability guarantee for this summary. We establish a stability theorem proving that our invariant is continuous with respect to the interleaving distance on multi-parameter modules. This result ensures robustness to noise and provides a theoretically sound, computationally tractable method for applying multi-parameter topological analysis to real-world data.
title Stable Barcodes for Multi-Parameter Persistent Homology
url https://doi.org/10.5281/zenodo.17681596