InfoFlow: A Transformer-based Time Series Anomaly Detection Model with Information Bottleneck and Normalizing Flows

InfoFlow: A Transformer-based Time Series Anomaly Detection Model with Information Bottleneck and Normalizing Flows | 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 Article InfoFlow: A Transformer-based Time Series Anomaly Detection Model with Information Bottleneck and Normalizing Flows Yuhua Mo, Hongzhu Fu, Sen Bai, Chao Deng, Tiantian Tang, Jian Lang, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4827149/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 Time series anomaly detection aims to identify unusual samples that significantly deviate from the majority of the data. The greatest challenge lies in endowing models with the ability to learn effective differentiation criteria. Previous methods have focused on learning intricate, sophisticated representations to overcome these difficulties. However, these approaches are often hindered by spurious correlations, which limit their potential for further improving model performance and causality.To solve the problem, we propose a novel approach, called InfoFlow, which accurately extracts and identifies pertinent details within sequences while excluding anomalous entries. This is achieved by utilizing information bottleneck constraints throughout a transformer-based model. Our primary discovery suggests that anomalous data are rare and unevenly distributed within the observed points, resulting in the phenomenon of anomaly concentration, which complicates the identification. To mitigate this issue, we employ normalizing flows to adjust the sequence distribution to a more uniform range. This approach accentuates subtle variations in the data and enhances the algorithm's sensitivity to anomalies, thereby facilitating easier detection of outlier points. Extensive experiments on real-world datasets demonstrate that our InfoFlowapproach outperforms state-of-the-art benchmarks. Physical sciences/Mathematics and computing/Computational science Physical sciences/Mathematics and computing/Information technology 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-4827149","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":349603714,"identity":"1c631d9f-ca7f-4e2e-be49-4b475495c2e8","order_by":0,"name":"Yuhua Mo","email":"","orcid":"","institution":"Internet Research Center, China Tobacco Guangxi Industrial Co., LTD.,China","correspondingAuthor":false,"prefix":"","firstName":"Yuhua","middleName":"","lastName":"Mo","suffix":""},{"id":349603715,"identity":"67615403-b882-4e73-a912-77c80c515c6b","order_by":1,"name":"Hongzhu Fu","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Hongzhu","middleName":"","lastName":"Fu","suffix":""},{"id":349603716,"identity":"e5f8b4fe-34f5-4278-b293-cec5b423015a","order_by":2,"name":"Sen Bai","email":"","orcid":"","institution":"Internet Research Center, China Tobacco Guangxi Industrial Co., LTD.,China","correspondingAuthor":false,"prefix":"","firstName":"Sen","middleName":"","lastName":"Bai","suffix":""},{"id":349603717,"identity":"06966c4c-37d3-44bc-98e1-2d52908d0e00","order_by":3,"name":"Chao Deng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6UlEQVRIiWNgGAWjYDCCA0DM2CAhx8beA+bz8BGpxcaYj+cMmMPDRqSWtMR5EjkQDkEtfMd7D7/4ueMwY5vk24OPP+bYybAxMD98dAOPFskz59Ise88cZmaTzks2OLgtGegwNmPjHDxaDG7kmBkzth1mY5POMZM4uI0ZqIUHyMan5f4bsBYeNskzIC31RGi5wWP8mLEtTYJNggek5TBhLZJncswYe9tsDNh4cowNzm47zsPGTMAvfMfPGH/42SZRP7/9jOGDym3V9vzszQ8f49MCBGwSqHxm/MrBSj4QVjMKRsEoGAUjGgAA+jBH9ox8+YQAAAAASUVORK5CYII=","orcid":"","institution":"Internet Research Center, China Tobacco Guangxi Industrial Co., LTD.,China","correspondingAuthor":true,"prefix":"","firstName":"Chao","middleName":"","lastName":"Deng","suffix":""},{"id":349603718,"identity":"6f5a58fa-269c-4a2c-8706-7a6ea4aeb139","order_by":4,"name":"Tiantian Tang","email":"","orcid":"","institution":"Internet Research Center, China Tobacco Guangxi Industrial Co., LTD.,China","correspondingAuthor":false,"prefix":"","firstName":"Tiantian","middleName":"","lastName":"Tang","suffix":""},{"id":349603719,"identity":"b5154cd5-42d3-4a8c-bd0e-ef87394e92e4","order_by":5,"name":"Jian Lang","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Jian","middleName":"","lastName":"Lang","suffix":""},{"id":349603720,"identity":"dfad6639-44b7-44b7-9e90-3535185eadb0","order_by":6,"name":"Fan Zhou","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Fan","middleName":"","lastName":"Zhou","suffix":""}],"badges":[],"createdAt":"2024-07-30 08:47:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4827149/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4827149/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":79668744,"identity":"83e18027-a5cb-4688-b478-bd6455eca124","added_by":"auto","created_at":"2025-04-01 10:46:50","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":484838,"visible":true,"origin":"","legend":"","description":"","filename":"2025SRHongzhuFu.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4827149/v1_covered_fbc83fc6-b798-4d45-8c5c-3deca346babe.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"InfoFlow: A Transformer-based Time Series Anomaly Detection Model with Information Bottleneck and Normalizing Flows","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":"","lastPublishedDoi":"10.21203/rs.3.rs-4827149/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4827149/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Time series anomaly detection aims to identify unusual samples that significantly deviate from the majority of the data. The greatest challenge lies in endowing models with the ability to learn effective differentiation criteria. Previous methods have focused on learning intricate, sophisticated representations to overcome these difficulties. However, these approaches are often hindered by spurious correlations, which limit their potential for further improving model performance and causality.To solve the problem, we propose a novel approach, called InfoFlow, which accurately extracts and identifies pertinent details within sequences while excluding anomalous entries. This is achieved by utilizing information bottleneck constraints throughout a transformer-based model. Our primary discovery suggests that anomalous data are rare and unevenly distributed within the observed points, resulting in the phenomenon of anomaly concentration, which complicates the identification. To mitigate this issue, we employ normalizing flows to adjust the sequence distribution to a more uniform range. This approach accentuates subtle variations in the data and enhances the algorithm's sensitivity to anomalies, thereby facilitating easier detection of outlier points. Extensive experiments on real-world datasets demonstrate that our InfoFlowapproach outperforms state-of-the-art benchmarks.","manuscriptTitle":"InfoFlow: A Transformer-based Time Series Anomaly Detection Model with Information Bottleneck and Normalizing Flows","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-05 03:55:28","doi":"10.21203/rs.3.rs-4827149/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":"318dcd1e-e8fc-4732-953e-4a38fc6bac6d","owner":[],"postedDate":"September 5th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":37107379,"name":"Physical sciences/Mathematics and computing/Computational science"},{"id":37107380,"name":"Physical sciences/Mathematics and computing/Information technology"}],"tags":[],"updatedAt":"2025-04-01T10:38:43+00:00","versionOfRecord":[],"versionCreatedAt":"2024-09-05 03:55:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4827149","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4827149","identity":"rs-4827149","version":["v1"]},"buildId":"_2-kVJe1T_tPrBINL-cwx","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}