DeepAI AI Chat
Log In Sign Up

On the Expressivity of Persistent Homology in Graph Learning

by   Bastian Rieck, et al.

Persistent homology, a technique from computational topology, has recently shown strong empirical performance in the context of graph classification. Being able to capture long range graph properties via higher-order topological features, such as cycles of arbitrary length, in combination with multi-scale topological descriptors, has improved predictive performance for data sets with prominent topological structures, such as molecules. At the same time, the theoretical properties of persistent homology have not been formally assessed in this context. This paper intends to bridge the gap between computational topology and graph machine learning by providing a brief introduction to persistent homology in the context of graphs, as well as a theoretical discussion and empirical analysis of its expressivity for graph learning tasks.


page 1

page 2

page 3

page 4


A Kernel for Multi-Parameter Persistent Homology

Topological data analysis and its main method, persistent homology, prov...

Topological Graph Neural Networks

Graph neural networks (GNNs) are a powerful architecture for tackling gr...

Persistent entropy: a scale-invariant topological statistic for analyzing cell arrangements

In this work, we explain how to use computational topology for detecting...

Graph Filtration Learning

We propose an approach to learning with graph-structured data in the pro...

Stable Homology-Based Cycle Centrality Measures for Weighted Graphs

Network centrality measures are concerned with evaluating the importance...

Long range actions, connectedness, and dismantlability in relational structures

In this paper we study alternative characterizations of dismantlability ...

Dory: Overcoming Barriers to Computing Persistent Homology

Persistent homology (PH) is an approach to topological data analysis (TD...