RTD-Lite: Scalable Topological Analysis for Comparing Weighted Graphs in Learning Tasks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tulchinskii, Eduard, Voronkova, Daria, Trofimov, Ilya, Burnaev, Evgeny, Barannikov, Serguei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912276769931264
author Tulchinskii, Eduard
Voronkova, Daria
Trofimov, Ilya
Burnaev, Evgeny
Barannikov, Serguei
author_facet Tulchinskii, Eduard
Voronkova, Daria
Trofimov, Ilya
Burnaev, Evgeny
Barannikov, Serguei
contents Topological methods for comparing weighted graphs are valuable in various learning tasks but often suffer from computational inefficiency on large datasets. We introduce RTD-Lite, a scalable algorithm that efficiently compares topological features, specifically connectivity or cluster structures at arbitrary scales, of two weighted graphs with one-to-one correspondence between vertices. Using minimal spanning trees in auxiliary graphs, RTD-Lite captures topological discrepancies with $O(n^2)$ time and memory complexity. This efficiency enables its application in tasks like dimensionality reduction and neural network training. Experiments on synthetic and real-world datasets demonstrate that RTD-Lite effectively identifies topological differences while significantly reducing computation time compared to existing methods. Moreover, integrating RTD-Lite into neural network training as a loss function component enhances the preservation of topological structures in learned representations. Our code is publicly available at https://github.com/ArGintum/RTD-Lite
format Preprint
id arxiv_https___arxiv_org_abs_2503_11910
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RTD-Lite: Scalable Topological Analysis for Comparing Weighted Graphs in Learning Tasks
Tulchinskii, Eduard
Voronkova, Daria
Trofimov, Ilya
Burnaev, Evgeny
Barannikov, Serguei
Machine Learning
Artificial Intelligence
Symplectic Geometry
Topological methods for comparing weighted graphs are valuable in various learning tasks but often suffer from computational inefficiency on large datasets. We introduce RTD-Lite, a scalable algorithm that efficiently compares topological features, specifically connectivity or cluster structures at arbitrary scales, of two weighted graphs with one-to-one correspondence between vertices. Using minimal spanning trees in auxiliary graphs, RTD-Lite captures topological discrepancies with $O(n^2)$ time and memory complexity. This efficiency enables its application in tasks like dimensionality reduction and neural network training. Experiments on synthetic and real-world datasets demonstrate that RTD-Lite effectively identifies topological differences while significantly reducing computation time compared to existing methods. Moreover, integrating RTD-Lite into neural network training as a loss function component enhances the preservation of topological structures in learned representations. Our code is publicly available at https://github.com/ArGintum/RTD-Lite
title RTD-Lite: Scalable Topological Analysis for Comparing Weighted Graphs in Learning Tasks
topic Machine Learning
Artificial Intelligence
Symplectic Geometry
url https://arxiv.org/abs/2503.11910