Positional Encoding meets Persistent Homology on Graphs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Verma, Yogesh, Souza, Amauri H., Garg, Vikas
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910991905718272
author Verma, Yogesh
Souza, Amauri H.
Garg, Vikas
author_facet Verma, Yogesh
Souza, Amauri H.
Garg, Vikas
contents The local inductive bias of message-passing graph neural networks (GNNs) hampers their ability to exploit key structural information (e.g., connectivity and cycles). Positional encoding (PE) and Persistent Homology (PH) have emerged as two promising approaches to mitigate this issue. PE schemes endow GNNs with location-aware features, while PH methods enhance GNNs with multiresolution topological features. However, a rigorous theoretical characterization of the relative merits and shortcomings of PE and PH has remained elusive. We bridge this gap by establishing that neither paradigm is more expressive than the other, providing novel constructions where one approach fails but the other succeeds. Our insights inform the design of a novel learnable method, PiPE (Persistence-informed Positional Encoding), which is provably more expressive than both PH and PE. PiPE demonstrates strong performance across a variety of tasks (e.g., molecule property prediction, graph classification, and out-of-distribution generalization), thereby advancing the frontiers of graph representation learning. Code is available at https://github.com/Aalto-QuML/PIPE.
format Preprint
id arxiv_https___arxiv_org_abs_2506_05814
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Positional Encoding meets Persistent Homology on Graphs
Verma, Yogesh
Souza, Amauri H.
Garg, Vikas
Machine Learning
Artificial Intelligence
Emerging Technologies
Social and Information Networks
The local inductive bias of message-passing graph neural networks (GNNs) hampers their ability to exploit key structural information (e.g., connectivity and cycles). Positional encoding (PE) and Persistent Homology (PH) have emerged as two promising approaches to mitigate this issue. PE schemes endow GNNs with location-aware features, while PH methods enhance GNNs with multiresolution topological features. However, a rigorous theoretical characterization of the relative merits and shortcomings of PE and PH has remained elusive. We bridge this gap by establishing that neither paradigm is more expressive than the other, providing novel constructions where one approach fails but the other succeeds. Our insights inform the design of a novel learnable method, PiPE (Persistence-informed Positional Encoding), which is provably more expressive than both PH and PE. PiPE demonstrates strong performance across a variety of tasks (e.g., molecule property prediction, graph classification, and out-of-distribution generalization), thereby advancing the frontiers of graph representation learning. Code is available at https://github.com/Aalto-QuML/PIPE.
title Positional Encoding meets Persistent Homology on Graphs
topic Machine Learning
Artificial Intelligence
Emerging Technologies
Social and Information Networks
url https://arxiv.org/abs/2506.05814