Graph Entropy Guided Node Embedding Dimension Selection for Graph Neural Networks
Graph representation learning has achieved great success in many areas, including e-commerce, chemistry, biology, etc. However, the fundamental problem of choosing the appropriate dimension of node embedding for a given graph still remains unsolved. The commonly used strategies for Node Embedding Di...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2021-08 |
---|---|
Hauptverfasser: | , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Luo, Gongxu Li, Jianxin Su, Jianlin Peng, Hao Yang, Carl Sun, Lichao Yu, Philip S He, Lifang |
description | Graph representation learning has achieved great success in many areas, including e-commerce, chemistry, biology, etc. However, the fundamental problem of choosing the appropriate dimension of node embedding for a given graph still remains unsolved. The commonly used strategies for Node Embedding Dimension Selection (NEDS) based on grid search or empirical knowledge suffer from heavy computation and poor model performance. In this paper, we revisit NEDS from the perspective of minimum entropy principle. Subsequently, we propose a novel Minimum Graph Entropy (MinGE) algorithm for NEDS with graph data. To be specific, MinGE considers both feature entropy and structure entropy on graphs, which are carefully designed according to the characteristics of the rich information in them. The feature entropy, which assumes the embeddings of adjacent nodes to be more similar, connects node features and link topology on graphs. The structure entropy takes the normalized degree as basic unit to further measure the higher-order structure of graphs. Based on them, we design MinGE to directly calculate the ideal node embedding dimension for any graph. Finally, comprehensive experiments with popular Graph Neural Networks (GNNs) on benchmark datasets demonstrate the effectiveness and generalizability of our proposed MinGE. |
format | Article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2524563370</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2524563370</sourcerecordid><originalsourceid>FETCH-proquest_journals_25245633703</originalsourceid><addsrcrecordid>eNqNi98KgjAcRkcQJOU7DLoW1ua0-zK7EqLuZbWfNdPN9ofo7TPqAbo6H5zvTFBEGVsl65TSGYqdawkhNMsp5yxCh9KK4YYL7a0ZXrgMSoLElZGAi_4MUip9xVvVg3bKaHyEDi7-sxpj8betIFjRjfBPY-9ugaaN6BzEP87RclecNvtksOYRwPm6NcHqUdWU05RnjOWE_fd6A2QpPuM</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2524563370</pqid></control><display><type>article</type><title>Graph Entropy Guided Node Embedding Dimension Selection for Graph Neural Networks</title><source>Free E- Journals</source><creator>Luo, Gongxu ; Li, Jianxin ; Su, Jianlin ; Peng, Hao ; Yang, Carl ; Sun, Lichao ; Yu, Philip S ; He, Lifang</creator><creatorcontrib>Luo, Gongxu ; Li, Jianxin ; Su, Jianlin ; Peng, Hao ; Yang, Carl ; Sun, Lichao ; Yu, Philip S ; He, Lifang</creatorcontrib><description>Graph representation learning has achieved great success in many areas, including e-commerce, chemistry, biology, etc. However, the fundamental problem of choosing the appropriate dimension of node embedding for a given graph still remains unsolved. The commonly used strategies for Node Embedding Dimension Selection (NEDS) based on grid search or empirical knowledge suffer from heavy computation and poor model performance. In this paper, we revisit NEDS from the perspective of minimum entropy principle. Subsequently, we propose a novel Minimum Graph Entropy (MinGE) algorithm for NEDS with graph data. To be specific, MinGE considers both feature entropy and structure entropy on graphs, which are carefully designed according to the characteristics of the rich information in them. The feature entropy, which assumes the embeddings of adjacent nodes to be more similar, connects node features and link topology on graphs. The structure entropy takes the normalized degree as basic unit to further measure the higher-order structure of graphs. Based on them, we design MinGE to directly calculate the ideal node embedding dimension for any graph. Finally, comprehensive experiments with popular Graph Neural Networks (GNNs) on benchmark datasets demonstrate the effectiveness and generalizability of our proposed MinGE.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Embedding ; Entropy ; Graph neural networks ; Graph representations ; Graphical representations ; Graphs ; Neural networks ; Nodes ; Topology</subject><ispartof>arXiv.org, 2021-08</ispartof><rights>2021. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (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><link.rule.ids>777,781</link.rule.ids></links><search><creatorcontrib>Luo, Gongxu</creatorcontrib><creatorcontrib>Li, Jianxin</creatorcontrib><creatorcontrib>Su, Jianlin</creatorcontrib><creatorcontrib>Peng, Hao</creatorcontrib><creatorcontrib>Yang, Carl</creatorcontrib><creatorcontrib>Sun, Lichao</creatorcontrib><creatorcontrib>Yu, Philip S</creatorcontrib><creatorcontrib>He, Lifang</creatorcontrib><title>Graph Entropy Guided Node Embedding Dimension Selection for Graph Neural Networks</title><title>arXiv.org</title><description>Graph representation learning has achieved great success in many areas, including e-commerce, chemistry, biology, etc. However, the fundamental problem of choosing the appropriate dimension of node embedding for a given graph still remains unsolved. The commonly used strategies for Node Embedding Dimension Selection (NEDS) based on grid search or empirical knowledge suffer from heavy computation and poor model performance. In this paper, we revisit NEDS from the perspective of minimum entropy principle. Subsequently, we propose a novel Minimum Graph Entropy (MinGE) algorithm for NEDS with graph data. To be specific, MinGE considers both feature entropy and structure entropy on graphs, which are carefully designed according to the characteristics of the rich information in them. The feature entropy, which assumes the embeddings of adjacent nodes to be more similar, connects node features and link topology on graphs. The structure entropy takes the normalized degree as basic unit to further measure the higher-order structure of graphs. Based on them, we design MinGE to directly calculate the ideal node embedding dimension for any graph. Finally, comprehensive experiments with popular Graph Neural Networks (GNNs) on benchmark datasets demonstrate the effectiveness and generalizability of our proposed MinGE.</description><subject>Algorithms</subject><subject>Embedding</subject><subject>Entropy</subject><subject>Graph neural networks</subject><subject>Graph representations</subject><subject>Graphical representations</subject><subject>Graphs</subject><subject>Neural networks</subject><subject>Nodes</subject><subject>Topology</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNi98KgjAcRkcQJOU7DLoW1ua0-zK7EqLuZbWfNdPN9ofo7TPqAbo6H5zvTFBEGVsl65TSGYqdawkhNMsp5yxCh9KK4YYL7a0ZXrgMSoLElZGAi_4MUip9xVvVg3bKaHyEDi7-sxpj8betIFjRjfBPY-9ugaaN6BzEP87RclecNvtksOYRwPm6NcHqUdWU05RnjOWE_fd6A2QpPuM</recordid><startdate>20210831</startdate><enddate>20210831</enddate><creator>Luo, Gongxu</creator><creator>Li, Jianxin</creator><creator>Su, Jianlin</creator><creator>Peng, Hao</creator><creator>Yang, Carl</creator><creator>Sun, Lichao</creator><creator>Yu, Philip S</creator><creator>He, Lifang</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>20210831</creationdate><title>Graph Entropy Guided Node Embedding Dimension Selection for Graph Neural Networks</title><author>Luo, Gongxu ; Li, Jianxin ; Su, Jianlin ; Peng, Hao ; Yang, Carl ; Sun, Lichao ; Yu, Philip S ; He, Lifang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_25245633703</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Algorithms</topic><topic>Embedding</topic><topic>Entropy</topic><topic>Graph neural networks</topic><topic>Graph representations</topic><topic>Graphical representations</topic><topic>Graphs</topic><topic>Neural networks</topic><topic>Nodes</topic><topic>Topology</topic><toplevel>online_resources</toplevel><creatorcontrib>Luo, Gongxu</creatorcontrib><creatorcontrib>Li, Jianxin</creatorcontrib><creatorcontrib>Su, Jianlin</creatorcontrib><creatorcontrib>Peng, Hao</creatorcontrib><creatorcontrib>Yang, Carl</creatorcontrib><creatorcontrib>Sun, Lichao</creatorcontrib><creatorcontrib>Yu, Philip S</creatorcontrib><creatorcontrib>He, Lifang</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</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>Luo, Gongxu</au><au>Li, Jianxin</au><au>Su, Jianlin</au><au>Peng, Hao</au><au>Yang, Carl</au><au>Sun, Lichao</au><au>Yu, Philip S</au><au>He, Lifang</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Graph Entropy Guided Node Embedding Dimension Selection for Graph Neural Networks</atitle><jtitle>arXiv.org</jtitle><date>2021-08-31</date><risdate>2021</risdate><eissn>2331-8422</eissn><abstract>Graph representation learning has achieved great success in many areas, including e-commerce, chemistry, biology, etc. However, the fundamental problem of choosing the appropriate dimension of node embedding for a given graph still remains unsolved. The commonly used strategies for Node Embedding Dimension Selection (NEDS) based on grid search or empirical knowledge suffer from heavy computation and poor model performance. In this paper, we revisit NEDS from the perspective of minimum entropy principle. Subsequently, we propose a novel Minimum Graph Entropy (MinGE) algorithm for NEDS with graph data. To be specific, MinGE considers both feature entropy and structure entropy on graphs, which are carefully designed according to the characteristics of the rich information in them. The feature entropy, which assumes the embeddings of adjacent nodes to be more similar, connects node features and link topology on graphs. The structure entropy takes the normalized degree as basic unit to further measure the higher-order structure of graphs. Based on them, we design MinGE to directly calculate the ideal node embedding dimension for any graph. Finally, comprehensive experiments with popular Graph Neural Networks (GNNs) on benchmark datasets demonstrate the effectiveness and generalizability of our proposed MinGE.</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, 2021-08 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_2524563370 |
source | Free E- Journals |
subjects | Algorithms Embedding Entropy Graph neural networks Graph representations Graphical representations Graphs Neural networks Nodes Topology |
title | Graph Entropy Guided Node Embedding Dimension Selection for Graph Neural Networks |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-20T03%3A36%3A15IST&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=Graph%20Entropy%20Guided%20Node%20Embedding%20Dimension%20Selection%20for%20Graph%20Neural%20Networks&rft.jtitle=arXiv.org&rft.au=Luo,%20Gongxu&rft.date=2021-08-31&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2524563370%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2524563370&rft_id=info:pmid/&rfr_iscdi=true |