Link Prediction via Adversarial Knowledge Distillation and Feature Aggregation

Link Prediction via Adversarial Knowledge Distillation and Feature Aggregation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Link Prediction via Adversarial Knowledge Distillation and Feature Aggregation Wen Li, Xiaoning Song, Wenjie Zhang, Yang Hua, Xiaojun Wu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4894235/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 31 Mar, 2025 Read the published version in Multimedia Systems → Version 1 posted 10 You are reading this latest preprint version Abstract Graph neural networks (GNN) have shown strong performance in link prediction tasks. However, it is susceptible to higher latency due to the trivial correlation of data in its neighborhood, which poses a challenge for its practical applica- tion. In contrast, although Multi-layer Perceptron (MLP) performs poorly, it has a shorter inference time and is more flexible in practical applications. We uti- lize a distillation model to combine the powerful inference capabilities of GNN with the inference effciency of MLP. Distillation models usually use a predefined distance function to quantify the differences between teacher-student networks, but this cannot be well applied to various complex scenarios. In addition, the limited node information severely affects the learning ability of MLP. Therefore, to cope with these problems. Firstly, we propose an Adversarial Generative Dis- criminator (AGD), which trains the discriminators and generators against each other to adaptively detect and reduce the differences. Secondly, we also propose the Feature Aggregation Module (FAM) to help the MLP obtain suffcient fea- ture information before distillation starts. In the experiments, it is shown that our approach can achieve good results in link prediction tasks, outperforming the baseline model Linkless Prediction (LLP) and maintaining a good inference speed on eight datasets in two different settings ∗ . ∗ The code on https://github.com/lwuen/LPVAKD.git Link Prediction Adversarial Knowledge Distillation Knowledge Graph Feature Aggregation Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 31 Mar, 2025 Read the published version in Multimedia Systems → Version 1 posted Editorial decision: Revision requested 24 Nov, 2024 Reviews received at journal 19 Nov, 2024 Reviewers agreed at journal 11 Nov, 2024 Reviews received at journal 17 Oct, 2024 Reviewers agreed at journal 18 Sep, 2024 Reviewers agreed at journal 18 Sep, 2024 Reviewers invited by journal 17 Sep, 2024 Editor assigned by journal 10 Sep, 2024 Submission checks completed at journal 12 Aug, 2024 First submitted to journal 11 Aug, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4894235","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":382020910,"identity":"87ed8328-99f8-4f5f-9a25-94af8f426f33","order_by":0,"name":"Wen Li","email":"","orcid":"","institution":"Jiangnan University","correspondingAuthor":false,"prefix":"","firstName":"Wen","middleName":"","lastName":"Li","suffix":""},{"id":382020915,"identity":"b12ba278-28f0-42fc-93d3-976b98b79c89","order_by":1,"name":"Xiaoning Song","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9klEQVRIiWNgGAWjYPACG2Y2BCeBKC1ppGs5jMwhoMXgAPvDxwW/zrPzsZ89/Lqg5g4DP3uOAcPPHbi1SDbwGBvP7LvNzMaTl2Y949gzBsmeNwaMvWdwa+Fn4GGT5u0BamHIMTPmYTvMYHAjx4CZsQ23FjYG9mdALeeY2fjfALX8O8xgT0gLPwODmTTPjwPMbBI5xo9524C2SBDQItkM9AtvQzJQyxszZt6+wzwSZ54VHOzFo8XgePvDxzx/7JLl+3OMP/N8OyzH35688cFPPFoYmIEY6IxkkL8kgAQPSPAAHg1Q8IfBDqT7A2GVo2AUjIJRMBIBAOyfRb0CRy5UAAAAAElFTkSuQmCC","orcid":"","institution":"Jiangnan University","correspondingAuthor":true,"prefix":"","firstName":"Xiaoning","middleName":"","lastName":"Song","suffix":""},{"id":382020918,"identity":"6a17e973-53c3-42c6-aee1-ad66f297094b","order_by":2,"name":"Wenjie Zhang","email":"","orcid":"","institution":"Jiangnan University","correspondingAuthor":false,"prefix":"","firstName":"Wenjie","middleName":"","lastName":"Zhang","suffix":""},{"id":382020922,"identity":"d4699a9f-9414-4e8f-82c2-6fc6597ae4aa","order_by":3,"name":"Yang Hua","email":"","orcid":"","institution":"Jiangnan University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Hua","suffix":""},{"id":382020923,"identity":"48531df5-d293-4ca7-8462-dc483c6b38ce","order_by":4,"name":"Xiaojun Wu","email":"","orcid":"","institution":"Jiangnan University","correspondingAuthor":false,"prefix":"","firstName":"Xiaojun","middleName":"","lastName":"Wu","suffix":""}],"badges":[],"createdAt":"2024-08-11 07:53:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4894235/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4894235/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00530-025-01733-5","type":"published","date":"2025-03-31T15:57:49+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80082402,"identity":"62aae24a-7e8b-4b3b-8348-280e276dd760","added_by":"auto","created_at":"2025-04-07 16:08:53","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1134332,"visible":true,"origin":"","legend":"","description":"","filename":"LinkPredictionviaAdversarialKnowledge.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4894235/v1_covered_34f6b37d-403c-4316-9907-edaa7c9f45e4.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Link Prediction via Adversarial Knowledge Distillation and Feature Aggregation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"multimedia-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mmsj","sideBox":"Learn more about [Multimedia Systems](http://link.springer.com/journal/530)","snPcode":"530","submissionUrl":"https://submission.nature.com/new-submission/530/3","title":"Multimedia Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Link Prediction, Adversarial Knowledge Distillation, Knowledge Graph, Feature Aggregation","lastPublishedDoi":"10.21203/rs.3.rs-4894235/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4894235/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGraph neural networks (GNN) have shown strong performance in link prediction tasks. However, it is susceptible to higher latency due to the trivial correlation of data in its neighborhood, which poses a challenge for its practical applica- tion. In contrast, although Multi-layer Perceptron (MLP) performs poorly, it has a shorter inference time and is more flexible in practical applications. We uti- lize a distillation model to combine the powerful inference capabilities of GNN with the inference effciency of MLP. Distillation models usually use a predefined distance function to quantify the differences between teacher-student networks, but this cannot be well applied to various complex scenarios. In addition, the limited node information severely affects the learning ability of MLP. Therefore, to cope with these problems. Firstly, we propose an Adversarial Generative Dis- criminator (AGD), which trains the discriminators and generators against each other to adaptively detect and reduce the differences. Secondly, we also propose the Feature Aggregation Module (FAM) to help the MLP obtain suffcient fea- ture information before distillation starts. In the experiments, it is shown that our approach can achieve good results in link prediction tasks, outperforming the baseline model Linkless Prediction (LLP) and maintaining a good inference speed on eight datasets in two different settings\u003csup\u003e∗\u003c/sup\u003e .\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e∗\u003c/sup\u003eThe code on https://github.com/lwuen/LPVAKD.git\u003c/p\u003e","manuscriptTitle":"Link Prediction via Adversarial Knowledge Distillation and Feature Aggregation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-12 03:30:13","doi":"10.21203/rs.3.rs-4894235/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-11-25T00:37:23+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-11-19T11:44:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"306785185146655386167249921182598005612","date":"2024-11-12T04:35:43+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-18T02:58:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"105072677542135203984735788790740963000","date":"2024-09-18T06:55:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"71734127495261351947832046966725311825","date":"2024-09-18T05:20:07+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-09-18T03:22:59+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-09-10T16:27:39+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-08-12T15:03:57+00:00","index":"","fulltext":""},{"type":"submitted","content":"Multimedia Systems","date":"2024-08-11T07:50:39+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"multimedia-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mmsj","sideBox":"Learn more about [Multimedia Systems](http://link.springer.com/journal/530)","snPcode":"530","submissionUrl":"https://submission.nature.com/new-submission/530/3","title":"Multimedia Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"8b4cbfd0-b6cf-40ae-a084-50e156d01259","owner":[],"postedDate":"December 12th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-04-07T16:07:01+00:00","versionOfRecord":{"articleIdentity":"rs-4894235","link":"https://doi.org/10.1007/s00530-025-01733-5","journal":{"identity":"multimedia-systems","isVorOnly":false,"title":"Multimedia Systems"},"publishedOn":"2025-03-31 15:57:49","publishedOnDateReadable":"March 31st, 2025"},"versionCreatedAt":"2024-12-12 03:30:13","video":"","vorDoi":"10.1007/s00530-025-01733-5","vorDoiUrl":"https://doi.org/10.1007/s00530-025-01733-5","workflowStages":[]},"version":"v1","identity":"rs-4894235","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4894235","identity":"rs-4894235","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}