Epitomic Variational Graph Autoencoder

Variational autoencoder (VAE) is a widely used generative model for learning latent representations. Burda et al. [3] in their seminal paper showed that learning capacity of VAE is limited by over-pruning. It is a phenomenon where a significant number of latent variables fail to capture any informat...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2020 25th International Conference on Pattern Recognition (ICPR) S. 7203 - 7210
Hauptverfasser: Khan, Rayyan Ahmad, Anwaar, Muhammad Umer, Kleinsteuber, Martin
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 10.01.2021
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Variational autoencoder (VAE) is a widely used generative model for learning latent representations. Burda et al. [3] in their seminal paper showed that learning capacity of VAE is limited by over-pruning. It is a phenomenon where a significant number of latent variables fail to capture any information about the input data and the corresponding hidden units become inactive. This adversely affects learning diverse and interpretable latent representations. As variational graph autoencoder (VGAE) extends VAE for graph-structured data, it inherits the over-pruning problem. In this paper, we adopt a model based approach and propose epitomic VGAE (EVGAE), a generative variational framework for graph datasets which successfully mitigates the over-pruning problem and also boosts the generative ability of VGAE. We consider EVGAE to consist of multiple sparse VGAE models, called epitomes, that are groups of latent variables sharing the latent space. This approach aids in increasing active units as epitomes compete to learn better representation of the graph data. We verify our claims via experiments on three benchmark datasets. Our experiments show that EVGAE has a better generative ability than VGAE. Moreover, EVGAE outperforms VGAE on link prediction task in citation networks.
AbstractList Variational autoencoder (VAE) is a widely used generative model for learning latent representations. Burda et al. [3] in their seminal paper showed that learning capacity of VAE is limited by over-pruning. It is a phenomenon where a significant number of latent variables fail to capture any information about the input data and the corresponding hidden units become inactive. This adversely affects learning diverse and interpretable latent representations. As variational graph autoencoder (VGAE) extends VAE for graph-structured data, it inherits the over-pruning problem. In this paper, we adopt a model based approach and propose epitomic VGAE (EVGAE), a generative variational framework for graph datasets which successfully mitigates the over-pruning problem and also boosts the generative ability of VGAE. We consider EVGAE to consist of multiple sparse VGAE models, called epitomes, that are groups of latent variables sharing the latent space. This approach aids in increasing active units as epitomes compete to learn better representation of the graph data. We verify our claims via experiments on three benchmark datasets. Our experiments show that EVGAE has a better generative ability than VGAE. Moreover, EVGAE outperforms VGAE on link prediction task in citation networks.
Author Anwaar, Muhammad Umer
Kleinsteuber, Martin
Khan, Rayyan Ahmad
Author_xml – sequence: 1
  givenname: Rayyan Ahmad
  surname: Khan
  fullname: Khan, Rayyan Ahmad
  email: rayyan.khan@tum.de
– sequence: 2
  givenname: Muhammad Umer
  surname: Anwaar
  fullname: Anwaar, Muhammad Umer
  email: umer.anwaar@tum.de
– sequence: 3
  givenname: Martin
  surname: Kleinsteuber
  fullname: Kleinsteuber, Martin
  email: kleinsteuber@tum.de
  organization: Mercateo AG,Munich,Germany
BookMark eNotjsFKw0AQQFfQg61-gSA5eUvc2cnuTo4l1LZQUEr1WibbCS602bDGg3-vYE_v8ODxZup6SIMo9Qi6AtDN86Z929VE2lVGG6iaGoxFuFIz8IbgTxDeqqflGKd0jqH44Bx5imngU7HKPH4Wi-8pyRDSUfKduun59CX3F87V-8ty367L7etq0y62ZTQapzIcgyXqHRHXWhBQa4--85bFETRiDHSOydRiLQlD6CmQb9B0tndeO5yrh_9uFJHDmOOZ88_hso6_G3A8pQ
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ICPR48806.2021.9412531
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 1728188083
9781728188089
EndPage 7210
ExternalDocumentID 9412531
Genre orig-research
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-i203t-cdc588f688a40e31300737b75ae6819e221b6a824e558ea1cf8c87932b5f67063
IEDL.DBID RIE
ISICitedReferencesCount 8
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000678409207046&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Thu Jun 29 18:39:16 EDT 2023
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-cdc588f688a40e31300737b75ae6819e221b6a824e558ea1cf8c87932b5f67063
PageCount 8
ParticipantIDs ieee_primary_9412531
PublicationCentury 2000
PublicationDate 2021-Jan.-10
PublicationDateYYYYMMDD 2021-01-10
PublicationDate_xml – month: 01
  year: 2021
  text: 2021-Jan.-10
  day: 10
PublicationDecade 2020
PublicationTitle 2020 25th International Conference on Pattern Recognition (ICPR)
PublicationTitleAbbrev ICPR
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
Score 2.1995685
Snippet Variational autoencoder (VAE) is a widely used generative model for learning latent representations. Burda et al. [3] in their seminal paper showed that...
SourceID ieee
SourceType Publisher
StartPage 7203
SubjectTerms Benchmark testing
EVGAE
Graph auto encoder
Graph neural networks
Over-pruning
Pattern recognition
Task analysis
VAE
Variational graph autoencoder
Title Epitomic Variational Graph Autoencoder
URI https://ieeexplore.ieee.org/document/9412531
WOSCitedRecordID wos000678409207046&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB5q8eBJpRXf7EE8ue0mu3nsUUqrXkoRld5Kkp3AHtyWduvvN7NdKoIXb0kIhBkyfJOZfDMAd1p4oYrExZmXPs4Yy2PDsjDNfeoKJaw0rmk2oaZTPZ_nsw487LkwiNh8PsMBDZtcfrF0WwqVDfMswDGRpg-UkjuuVkv6ZUk-fBnNXuk60scDzgbt5l9dUxrQmBz_77gT6P-w76LZHldOoYNVD-7Hq2B7n6WLPsLrto3gRU9Ubjp63NZLqkdZ4LoP75Px2-g5bnscxCVP0jp2hRNae6m1yRJMKbmkUmWVMCgDWCPnLKhL8wyF0GiY89rpYFPcCi9V8C_OoFstKzyHKJXojFPGcku8extcN048d5OLQjFrLqBHMi5WuzIWi1a8y7-Xr-CI1EjRBpZcQ7deb_EGDt1XXW7Wt43uvwFy5oVP
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dSwMxDA9jCvqksonf3oP45G3X3vXae5SxueEcQ6bsbbS9FPawD-bNv9_mPCaCL761pVCSNiRN8ksA7pRwQuaRDROXujBhLAs1S_w0c7HNpTCptmWzCTkaqek0G9fgYYeFQcQy-QxbNCxj-fnKbslV1s4Sr44JNL1HnbMqtFYF-2VR1h50xq_0ICn1gLNWtf1X35RSbfSO_nfgMTR_8HfBeKdZTqCGywbcd9de-hZzG7z7_23lwwueqOB08LgtVlSRMsdNE9563UmnH1ZdDsI5j-IitLkVSrlUKZ1EGFN4ScbSSKEx9eoaOWeeYYonKIRCzaxTVnmp4ka4VHoL4xTqy9USzyCIU7TaSm24IeS98cYbJ6S7zkQumdHn0CAaZ-vvQhaziryLv5dv4aA_eRnOhoPR8yUcEkvJ98CiK6gXmy1ew779LOYfm5vyHr4AmxaImA
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2020+25th+International+Conference+on+Pattern+Recognition+%28ICPR%29&rft.atitle=Epitomic+Variational+Graph+Autoencoder&rft.au=Khan%2C+Rayyan+Ahmad&rft.au=Anwaar%2C+Muhammad+Umer&rft.au=Kleinsteuber%2C+Martin&rft.date=2021-01-10&rft.pub=IEEE&rft.spage=7203&rft.epage=7210&rft_id=info:doi/10.1109%2FICPR48806.2021.9412531&rft.externalDocID=9412531