Salvato in:
Dettagli Bibliografici
Autori principali: Fiorini, Stefano, Coniglio, Stefano, Ciavotta, Michele, Messina, Enza
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:https://arxiv.org/abs/2312.17361
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911743241879552
author Fiorini, Stefano
Coniglio, Stefano
Ciavotta, Michele
Messina, Enza
author_facet Fiorini, Stefano
Coniglio, Stefano
Ciavotta, Michele
Messina, Enza
contents We introduce QuaterGCN, a spectral Graph Convolutional Network (GCN) with quaternion-valued weights at whose core lies the Quaternionic Laplacian, a quaternion-valued Laplacian matrix by whose proposal we generalize two widely-used Laplacian matrices: the classical Laplacian (defined for undirected graphs) and the complex-valued Sign-Magnetic Laplacian (proposed to handle digraphs with weights of arbitrary sign). In addition to its generality, our Quaternionic Laplacian is the only Laplacian to completely preserve the topology of a digraph, as it can handle graphs and digraphs containing antiparallel pairs of edges (digons) of different weights without reducing them to a single (directed or undirected) edge as done with other Laplacians. Experimental results show the superior performance of QuaterGCN compared to other state-of-the-art GCNs, particularly in scenarios where the information the digons carry is crucial to successfully address the task at hand.
format Preprint
id arxiv_https___arxiv_org_abs_2312_17361
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Graph Learning in 4D: a Quaternion-valued Laplacian to Enhance Spectral GCNs
Fiorini, Stefano
Coniglio, Stefano
Ciavotta, Michele
Messina, Enza
Machine Learning
We introduce QuaterGCN, a spectral Graph Convolutional Network (GCN) with quaternion-valued weights at whose core lies the Quaternionic Laplacian, a quaternion-valued Laplacian matrix by whose proposal we generalize two widely-used Laplacian matrices: the classical Laplacian (defined for undirected graphs) and the complex-valued Sign-Magnetic Laplacian (proposed to handle digraphs with weights of arbitrary sign). In addition to its generality, our Quaternionic Laplacian is the only Laplacian to completely preserve the topology of a digraph, as it can handle graphs and digraphs containing antiparallel pairs of edges (digons) of different weights without reducing them to a single (directed or undirected) edge as done with other Laplacians. Experimental results show the superior performance of QuaterGCN compared to other state-of-the-art GCNs, particularly in scenarios where the information the digons carry is crucial to successfully address the task at hand.
title Graph Learning in 4D: a Quaternion-valued Laplacian to Enhance Spectral GCNs
topic Machine Learning
url https://arxiv.org/abs/2312.17361