Robust graph representation learning via out-of-distribution detection approach

Robust graph representation learning via out-of-distribution detection approach | 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 Robust graph representation learning via out-of-distribution detection approach Esmaeil Bastami, Hadi Soltanizadeh, Mohammad Rahmanimanesh, Parviz Keshavarzi, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3887563/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Graph neural networks (GNNs) are powerful models capable of learning from graph-structured data and performing various tasks. GNNs are susceptible to poisoning attacks, in which sophisticated attackers inject malicious nodes or edges into the graph topology to degrade their performance. The existing defense mechanisms, such as adversarial training, are ineffective in improving the robustness of GNN models. Fake nodes can be utilized creatively to deceive traditional GNN neural functions. In this paper, we propose a robust GNN model empowered by a reliable aggregation function based on the OOD detection concept and a novel method. A key idea of RGRO is to train a model that maps the nodes to a latent space in which we can measure the distance between the nodes and their distribution. The Mahalanobis distance is proposed as a superior alternative to cosine distance in order to exploit the homophily rule better and to consider the contextual information of the nodes. The Mahalanobis distance, which considers the covariance of the data and is scale-invariant, allows for a more effective application of the homophily rule. Mahalanobis distance is explored in this study to enhance the robustness and accuracy of outlier detection in graph data, thus contributing to a better understanding of homophily. The RGRO model can improve accuracy by removing poisoned data without knowing any types of poisoning attacks or GNN algorithms. RGRO is evaluated with four typical defense strategies based on two types of poisoning attacks implemented on different realistic datasets. Based on the results, RGRO can detect poisoned data against attacks effectively and efficiently. In the best scenario, RGRO improves the accuracy of the GNN model by 0.86. Adversarial Attack Out-of-distribution Graph Neural Networks Robust aggregation function Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-3887563","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":268668694,"identity":"a967012d-66b9-480b-a380-0a678e6895b1","order_by":0,"name":"Esmaeil Bastami","email":"","orcid":"","institution":"Semnan University","correspondingAuthor":false,"prefix":"","firstName":"Esmaeil","middleName":"","lastName":"Bastami","suffix":""},{"id":268668695,"identity":"c68e3c4a-6e74-458a-af13-1b652d61b248","order_by":1,"name":"Hadi Soltanizadeh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABE0lEQVRIiWNgGAWjYBACAxDxAIjZgFiCscEGSY4Zj5aEBLiWNBK0MEC0HCbsMHMG9ocPEn/Y5fHxH3544+OO8/b8YmcPfi5gsJNnYOd9gE2LZQOPsUFCQnIxG8MxY8uZZ24nzpydlyw9gyHZsIGZ3QCrww7wsEkkJDAntjE2mEnztt1OMLidYyDNw8CcwMDMht0vB9ifAbXUJ7Yxs3+T/tt2zt7+do7xbx6GejxaGMyAWg4ntrHxmEkzth1g3CCdYwa05TAeLSC/pB1PbOPhKbbsbUtOnHE7x8yax+C4YRtuhz188MGmOnF+//GNN3622dnzz84xvs1TUS3Pz38MqxYG+QfYjYJE7igYBaNgFIwC8gAAFsZTwGR6bVYAAAAASUVORK5CYII=","orcid":"","institution":"Semnan University","correspondingAuthor":true,"prefix":"","firstName":"Hadi","middleName":"","lastName":"Soltanizadeh","suffix":""},{"id":268668696,"identity":"0f7fe978-f62e-46be-a15f-4609d812828c","order_by":2,"name":"Mohammad Rahmanimanesh","email":"","orcid":"","institution":"Semnan University","correspondingAuthor":false,"prefix":"","firstName":"Mohammad","middleName":"","lastName":"Rahmanimanesh","suffix":""},{"id":268668697,"identity":"9be8ac94-ca2f-4471-88b5-2993a4f5f96f","order_by":3,"name":"Parviz Keshavarzi","email":"","orcid":"","institution":"Semnan University","correspondingAuthor":false,"prefix":"","firstName":"Parviz","middleName":"","lastName":"Keshavarzi","suffix":""},{"id":268668698,"identity":"db4b0698-b2d2-47df-bcbe-285f92da1aa5","order_by":4,"name":"Masoud Asadpour","email":"","orcid":"","institution":"Tehran University","correspondingAuthor":false,"prefix":"","firstName":"Masoud","middleName":"","lastName":"Asadpour","suffix":""}],"badges":[],"createdAt":"2024-01-22 09:45:46","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3887563/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3887563/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":54751487,"identity":"7efb5a42-d0ef-4a74-9cf3-f7acc8bc62d4","added_by":"auto","created_at":"2024-04-16 08:44:28","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1017661,"visible":true,"origin":"","legend":"","description":"","filename":"021029.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3887563/v1_covered_b03804d2-164d-4eae-b8fe-440e13973cd6.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Robust graph representation learning via out-of-distribution detection approach","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Adversarial Attack, Out-of-distribution, Graph Neural Networks, Robust aggregation function","lastPublishedDoi":"10.21203/rs.3.rs-3887563/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3887563/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGraph neural networks (GNNs) are powerful models capable of learning from graph-structured data and performing various tasks. GNNs are susceptible to poisoning attacks, in which sophisticated attackers inject malicious nodes or edges into the graph topology to degrade their performance. The existing defense mechanisms, such as adversarial training, are ineffective in improving the robustness of GNN models. Fake nodes can be utilized creatively to deceive traditional GNN neural functions. In this paper, we propose a robust GNN model empowered by a reliable aggregation function based on the OOD detection concept and a novel method. A key idea of RGRO is to train a model that maps the nodes to a latent space in which we can measure the distance between the nodes and their distribution. The Mahalanobis distance is proposed as a superior alternative to cosine distance in order to exploit the homophily rule better and to consider the contextual information of the nodes. The Mahalanobis distance, which considers the covariance of the data and is scale-invariant, allows for a more effective application of the homophily rule. Mahalanobis distance is explored in this study to enhance the robustness and accuracy of outlier detection in graph data, thus contributing to a better understanding of homophily. The RGRO model can improve accuracy by removing poisoned data without knowing any types of poisoning attacks or GNN algorithms. RGRO is evaluated with four typical defense strategies based on two types of poisoning attacks implemented on different realistic datasets. Based on the results, RGRO can detect poisoned data against attacks effectively and efficiently. In the best scenario, RGRO improves the accuracy of the GNN model by 0.86.\u003c/p\u003e","manuscriptTitle":"Robust graph representation learning via out-of-distribution detection approach","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-24 18:32:52","doi":"10.21203/rs.3.rs-3887563/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ce2d980f-a56a-4842-93f8-27949083aac5","owner":[],"postedDate":"January 24th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-04-16T08:36:19+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-24 18:32:52","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3887563","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3887563","identity":"rs-3887563","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}