Projections of Model Spaces for Latent Graph Inference

Graph Neural Networks leverage the connectivity structure of graphs as an inductive bias. Latent graph inference focuses on learning an adequate graph structure to diffuse information on and improve the downstream performance of the model. In this work we employ stereographic projections of the hype...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-04
Main Authors: Haitz Sáez de Ocáriz Borde, Arroyo, Álvaro, Posner, Ingmar
Format: Article
Language:English
Subjects:
Graph neural networks
Graph theory
Graphs
Inference
Riemann manifold
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Haitz Sáez de Ocáriz Borde
Arroyo, Álvaro
Posner, Ingmar
description Graph Neural Networks leverage the connectivity structure of graphs as an inductive bias. Latent graph inference focuses on learning an adequate graph structure to diffuse information on and improve the downstream performance of the model. In this work we employ stereographic projections of the hyperbolic and spherical model spaces, as well as products of Riemannian manifolds, for the purpose of latent graph inference. Stereographically projected model spaces achieve comparable performance to their non-projected counterparts, while providing theoretical guarantees that avoid divergence of the spaces when the curvature tends to zero. We perform experiments on both homophilic and heterophilic graphs.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2789557422</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2789557422</sourcerecordid><originalsourceid>FETCH-proquest_journals_27895574223</originalsourceid><addsrcrecordid>eNqNjUsKwjAUAIMgWLR3eOC6EBPT1LX4AwXB7kuIL2gpeTFJ728XHsDVLGZgZqwQUm6qZivEgpUp9ZxzUWuhlCxYfY_Uo81v8gnIwY2eOMAjGIsJHEW4mow-wyma8IKLdxjRW1yxuTNDwvLHJVsfD-3-XIVInxFT7noao59UJ3SzU0pPd_lf9QWWJjTm</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2789557422</pqid></control><display><type>article</type><title>Projections of Model Spaces for Latent Graph Inference</title><source>Publicly Available Content Database</source><creator>Haitz Sáez de Ocáriz Borde ; Arroyo, Álvaro ; Posner, Ingmar</creator><creatorcontrib>Haitz Sáez de Ocáriz Borde ; Arroyo, Álvaro ; Posner, Ingmar</creatorcontrib><description>Graph Neural Networks leverage the connectivity structure of graphs as an inductive bias. Latent graph inference focuses on learning an adequate graph structure to diffuse information on and improve the downstream performance of the model. In this work we employ stereographic projections of the hyperbolic and spherical model spaces, as well as products of Riemannian manifolds, for the purpose of latent graph inference. Stereographically projected model spaces achieve comparable performance to their non-projected counterparts, while providing theoretical guarantees that avoid divergence of the spaces when the curvature tends to zero. We perform experiments on both homophilic and heterophilic graphs.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Graph neural networks ; Graph theory ; Graphs ; Inference ; Riemann manifold</subject><ispartof>arXiv.org, 2023-04</ispartof><rights>2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2789557422?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>776,780,25731,36989,44566</link.rule.ids></links><search><creatorcontrib>Haitz Sáez de Ocáriz Borde</creatorcontrib><creatorcontrib>Arroyo, Álvaro</creatorcontrib><creatorcontrib>Posner, Ingmar</creatorcontrib><title>Projections of Model Spaces for Latent Graph Inference</title><title>arXiv.org</title><description>Graph Neural Networks leverage the connectivity structure of graphs as an inductive bias. Latent graph inference focuses on learning an adequate graph structure to diffuse information on and improve the downstream performance of the model. In this work we employ stereographic projections of the hyperbolic and spherical model spaces, as well as products of Riemannian manifolds, for the purpose of latent graph inference. Stereographically projected model spaces achieve comparable performance to their non-projected counterparts, while providing theoretical guarantees that avoid divergence of the spaces when the curvature tends to zero. We perform experiments on both homophilic and heterophilic graphs.</description><subject>Graph neural networks</subject><subject>Graph theory</subject><subject>Graphs</subject><subject>Inference</subject><subject>Riemann manifold</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNqNjUsKwjAUAIMgWLR3eOC6EBPT1LX4AwXB7kuIL2gpeTFJ728XHsDVLGZgZqwQUm6qZivEgpUp9ZxzUWuhlCxYfY_Uo81v8gnIwY2eOMAjGIsJHEW4mow-wyma8IKLdxjRW1yxuTNDwvLHJVsfD-3-XIVInxFT7noao59UJ3SzU0pPd_lf9QWWJjTm</recordid><startdate>20230412</startdate><enddate>20230412</enddate><creator>Haitz Sáez de Ocáriz Borde</creator><creator>Arroyo, Álvaro</creator><creator>Posner, Ingmar</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20230412</creationdate><title>Projections of Model Spaces for Latent Graph Inference</title><author>Haitz Sáez de Ocáriz Borde ; Arroyo, Álvaro ; Posner, Ingmar</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_27895574223</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Graph neural networks</topic><topic>Graph theory</topic><topic>Graphs</topic><topic>Inference</topic><topic>Riemann manifold</topic><toplevel>online_resources</toplevel><creatorcontrib>Haitz Sáez de Ocáriz Borde</creatorcontrib><creatorcontrib>Arroyo, Álvaro</creatorcontrib><creatorcontrib>Posner, Ingmar</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Haitz Sáez de Ocáriz Borde</au><au>Arroyo, Álvaro</au><au>Posner, Ingmar</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Projections of Model Spaces for Latent Graph Inference</atitle><jtitle>arXiv.org</jtitle><date>2023-04-12</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>Graph Neural Networks leverage the connectivity structure of graphs as an inductive bias. Latent graph inference focuses on learning an adequate graph structure to diffuse information on and improve the downstream performance of the model. In this work we employ stereographic projections of the hyperbolic and spherical model spaces, as well as products of Riemannian manifolds, for the purpose of latent graph inference. Stereographically projected model spaces achieve comparable performance to their non-projected counterparts, while providing theoretical guarantees that avoid divergence of the spaces when the curvature tends to zero. We perform experiments on both homophilic and heterophilic graphs.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-04
issn 2331-8422
language eng
recordid cdi_proquest_journals_2789557422
source Publicly Available Content Database
subjects Graph neural networks
Graph theory
Graphs
Inference
Riemann manifold
title Projections of Model Spaces for Latent Graph Inference
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-06T20%3A14%3A38IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Projections%20of%20Model%20Spaces%20for%20Latent%20Graph%20Inference&rft.jtitle=arXiv.org&rft.au=Haitz%20S%C3%A1ez%20de%20Oc%C3%A1riz%20Borde&rft.date=2023-04-12&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2789557422%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_27895574223%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2789557422&rft_id=info:pmid/&rfr_iscdi=true