Multimodal deep learning framework for shadowbanking risk prediction - dynamic decisionoptimization integrating knowledge graph andreinforcement learning

Multimodal deep learning framework for shadowbanking risk prediction - dynamic decisionoptimization integrating knowledge graph andreinforcement learning | 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 Multimodal deep learning framework for shadowbanking risk prediction - dynamic decisionoptimization integrating knowledge graph andreinforcement learning Tong Qin This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7351508/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Amid increasing digitization and globalization of financial systems, the detection and mitigation of systemic risk withinnon-traditional financial sectors has emerged as a critical research imperative in computer science. Traditional statistical andeconometric models for risk assessment often suffer from static assumptions, limited capacity to model interdependencies, andlack of regulatory interpretability—shortcomings that hinder real-time and scalable solutions in complex financial ecosystems.To overcome these limitations, we propose a multimodal deep learning framework that integrates a graph-theoretic neuralarchitecture, GFA-Net, with a policy-aware strategic module, PCS-Flow. GFA-Net encodes financial systems as dynamictransaction graphs enriched with semantic and regulatory features, enabling robust structural learning and forward simulationacross accounting periods. PCS-Flow further ensures that model outputs remain consistent under heterogeneous policyregimes and evolving fiscal scenarios incorporating differentiable scenario perturbations and compliance regularizers.Through these synergistic components, our approach delivers a unified solution for forecasting, anomaly detection, anddecision optimization in high-dimensional financial environments. Experimental results on simulated and real-world datasetsdemonstrate superior accuracy, compliance fidelity, and temporal stability, thus validating the utility of our method forpolicy-consistent risk prediction. This work contributes to the field by advancing interpretable, regulation-aware machinelearning frameworks capable of navigating the evolving landscape of financial technologies. Physical sciences/Engineering Physical sciences/Mathematics and computing Graph Neural Networks Financial Compliance Multimodal Learning Dynamic Systems Policy-Aware Optimization Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 23 Oct, 2025 Reviews received at journal 22 Oct, 2025 Reviews received at journal 03 Oct, 2025 Reviewers agreed at journal 03 Oct, 2025 Reviewers agreed at journal 02 Oct, 2025 Reviewers invited by journal 01 Oct, 2025 Editor invited by journal 04 Sep, 2025 Editor assigned by journal 13 Aug, 2025 Submission checks completed at journal 12 Aug, 2025 First submitted to journal 12 Aug, 2025 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-7351508","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":528933313,"identity":"a70ab8a9-3672-4cc6-adf4-7181946af20b","order_by":0,"name":"Tong Qin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4klEQVRIie3PMQrCMBiG4b8E0uVX15SCZ4gUrELAq1SETA6OblY6dKm7UPAM3sCWgC4eoOKkvUDHDg4G3ERa3Rzybgl5CB+AyfSP2QAV1GJ1iEN9WgCwNoEErK2VSGuPGUDGvyVAlbVnwZdkQjp5WKEkvlMe3YqLvhOS271o/KU7XW9Hgo5TKVnGpecC9bx5I0G/RJQI1/lQEzXdAVK3hQyiB1UMLudfiJ7PeYEvkrYSpbdsEhk4iZyNznqLE7VsseNNHta1CHq2yovlUvTZKbqVTUT/835BGp9/JCaTyWR66wlaBEZT70/1wQAAAABJRU5ErkJggg==","orcid":"","institution":"Chongqing University","correspondingAuthor":true,"prefix":"","firstName":"Tong","middleName":"","lastName":"Qin","suffix":""}],"badges":[],"createdAt":"2025-08-12 05:23:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7351508/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7351508/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93458409,"identity":"d4214d5c-51b9-4227-8c45-7d73b39b1112","added_by":"auto","created_at":"2025-10-14 05:50:48","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3919,"visible":true,"origin":"","legend":"","description":"","filename":"fa0ccc2789804f9b92c6a4f128aa6e7c.json","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/ab26ba7ee8a715456b352812.json"},{"id":93458133,"identity":"14223bfa-2f40-4965-b90a-ba3eb3d3ab2d","added_by":"auto","created_at":"2025-10-14 05:42:48","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":133971,"visible":true,"origin":"","legend":"","description":"","filename":"Fig1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/84207a899c388c835e29b453.pdf"},{"id":93460047,"identity":"903364e5-530a-4e87-b5eb-9e8cbf79e91e","added_by":"auto","created_at":"2025-10-14 06:06:48","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":93084,"visible":true,"origin":"","legend":"","description":"","filename":"Fig2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/070bae67cebc2d7a33c6d5b6.pdf"},{"id":93459745,"identity":"48ba6ec6-5cfb-4b3e-846e-556c5f1e0bdf","added_by":"auto","created_at":"2025-10-14 05:58:48","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":390541,"visible":true,"origin":"","legend":"","description":"","filename":"Fig3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/e74e1f59e8eb5410a807218c.pdf"},{"id":93458135,"identity":"28147328-19f5-47d3-870d-1f96eb594a50","added_by":"auto","created_at":"2025-10-14 05:42:48","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":112125,"visible":true,"origin":"","legend":"","description":"","filename":"Fig4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/c0b8db455fde9eb6bd314014.pdf"},{"id":93460048,"identity":"92069d9d-044b-47ad-a0a1-d7cc0a973166","added_by":"auto","created_at":"2025-10-14 06:06:48","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2054514,"visible":true,"origin":"","legend":"","description":"","filename":"ScientificReports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/4b9a1e5f1d7d02a3d91d9217.pdf"},{"id":93458143,"identity":"bd3955df-6ab1-47c5-a3da-61c552245eb5","added_by":"auto","created_at":"2025-10-14 05:42:48","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":361624,"visible":true,"origin":"","legend":"","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/70842a650560b7c6a79d9be7.png"},{"id":93458410,"identity":"da882431-d743-4477-b2b4-a97058eda913","added_by":"auto","created_at":"2025-10-14 05:50:48","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":224592,"visible":true,"origin":"","legend":"","description":"","filename":"fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/004dd6891204b42f5c97c04c.png"},{"id":93458412,"identity":"8bbf8ae0-009c-49ed-aac7-bceedeba1d71","added_by":"auto","created_at":"2025-10-14 05:50:48","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":509183,"visible":true,"origin":"","legend":"","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/f8977e25de11a9249081fd0b.png"},{"id":93458140,"identity":"15b6f68b-71ba-4f97-8860-1137a7ba8d74","added_by":"auto","created_at":"2025-10-14 05:42:48","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":227264,"visible":true,"origin":"","legend":"","description":"","filename":"fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/0cf73baa5adfe87491e77af7.png"},{"id":93458147,"identity":"bad513bc-04c1-4b80-a332-67fdd0df9ae6","added_by":"auto","created_at":"2025-10-14 05:42:48","extension":"ldf","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49107,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrvltwaall.ldf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/5bbf70ff7cd1c1ce3aedb1cd.ldf"},{"id":93458151,"identity":"6879d607-086a-4df5-9aa3-91eacde56f6a","added_by":"auto","created_at":"2025-10-14 05:42:49","extension":"ldf","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":268657,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrvltwaen.ldf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/d5807f84005676025a05e845.ldf"},{"id":93459746,"identity":"175d3418-b40d-40f0-b0e4-04c454c49cfe","added_by":"auto","created_at":"2025-10-14 05:58:48","extension":"sty","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":15480,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrv.sty","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/9a5d459517c5d7bd4125ecaa.sty"},{"id":93458150,"identity":"b2046a54-0e6e-4bfd-97b7-65706bea025c","added_by":"auto","created_at":"2025-10-14 05:42:49","extension":"aux","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12444,"visible":true,"origin":"","legend":"","description":"","filename":"main.aux","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/dfee99487f6df7fffbbfb066.aux"},{"id":93458417,"identity":"37174b72-bd04-4de6-a043-4c36ca19ab8f","added_by":"auto","created_at":"2025-10-14 05:50:48","extension":"bbl","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11451,"visible":true,"origin":"","legend":"","description":"","filename":"main.bbl","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/0853d7d60d5a4191bc611445.bbl"},{"id":93460049,"identity":"469ec46d-fcef-4efb-8755-e5d1d71bf135","added_by":"auto","created_at":"2025-10-14 06:06:48","extension":"blg","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":978,"visible":true,"origin":"","legend":"","description":"","filename":"main.blg","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/e46b4e46de804bf7d00f1c7a.blg"},{"id":93458418,"identity":"62fd8e7e-572f-4773-b513-9ed55fb51a6c","added_by":"auto","created_at":"2025-10-14 05:50:48","extension":"log","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41827,"visible":true,"origin":"","legend":"","description":"","filename":"main.log","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/3b92a426aac81caf64fafae0.log"},{"id":93458136,"identity":"3e6f2c6a-2d1b-4ecb-abd3-575b30a64d8a","added_by":"auto","created_at":"2025-10-14 05:42:48","extension":"out","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2944,"visible":true,"origin":"","legend":"","description":"","filename":"main.out","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/e480f95715d282dac518c4d7.out"},{"id":93459750,"identity":"4e9e3ca2-b57f-4209-a416-27ad35738d24","added_by":"auto","created_at":"2025-10-14 05:58:48","extension":"pdf","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1940936,"visible":true,"origin":"","legend":"","description":"","filename":"main.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/71677a0063ee41dda51ba5b5.pdf"},{"id":93458153,"identity":"dade72af-636d-4452-8a56-70001b15ab26","added_by":"auto","created_at":"2025-10-14 05:42:49","extension":"gz","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":186358,"visible":true,"origin":"","legend":"","description":"","filename":"main.synctex.gz","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/2702fa23ad01b2a15240a5ae.gz"},{"id":93458420,"identity":"492a7ca5-602a-40ea-b780-74fdb79c1924","added_by":"auto","created_at":"2025-10-14 05:50:49","extension":"bst","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36153,"visible":true,"origin":"","legend":"","description":"","filename":"naturemagdoi.bst","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/e672ef149e80514744515431.bst"},{"id":93458155,"identity":"6e2ad92a-8343-40f9-992d-60d75bd84c9d","added_by":"auto","created_at":"2025-10-14 05:42:49","extension":"jpg","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":379752,"visible":true,"origin":"","legend":"","description":"","filename":"stream.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/cbf8ab28d721f1c6e6dad784.jpg"},{"id":93458152,"identity":"3ca1fd0d-9389-47d5-a35a-c1046920356e","added_by":"auto","created_at":"2025-10-14 05:42:49","extension":"cls","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5824,"visible":true,"origin":"","legend":"","description":"","filename":"wlscirep.cls","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/681e11e5ac7105237b038181.cls"},{"id":93458421,"identity":"06d2a63d-4651-4e8a-b55d-5a1ffe212075","added_by":"auto","created_at":"2025-10-14 05:50:49","extension":"xml","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":141528,"visible":true,"origin":"","legend":"","description":"","filename":"fa0ccc2789804f9b92c6a4f128aa6e7c1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1/9cf8b9b456324023100c7d0f.xml"},{"id":93460696,"identity":"15e96356-b6dc-4109-8138-c73213be82c6","added_by":"auto","created_at":"2025-10-14 06:14:52","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2213569,"visible":true,"origin":"","legend":"","description":"","filename":"ScientificReports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7351508/v1_covered_56e1eb82-c6c8-4126-923c-8b08ce72f78d.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Multimodal deep learning framework for shadowbanking risk prediction - dynamic decisionoptimization integrating knowledge graph andreinforcement learning","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Graph Neural Networks, Financial Compliance, Multimodal Learning, Dynamic Systems, Policy-Aware Optimization","lastPublishedDoi":"10.21203/rs.3.rs-7351508/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7351508/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Amid increasing digitization and globalization of financial systems, the detection and mitigation of systemic risk withinnon-traditional financial sectors has emerged as a critical research imperative in computer science. Traditional statistical andeconometric models for risk assessment often suffer from static assumptions, limited capacity to model interdependencies, andlack of regulatory interpretability—shortcomings that hinder real-time and scalable solutions in complex financial ecosystems.To overcome these limitations, we propose a multimodal deep learning framework that integrates a graph-theoretic neuralarchitecture, GFA-Net, with a policy-aware strategic module, PCS-Flow. GFA-Net encodes financial systems as dynamictransaction graphs enriched with semantic and regulatory features, enabling robust structural learning and forward simulationacross accounting periods. PCS-Flow further ensures that model outputs remain consistent under heterogeneous policyregimes and evolving fiscal scenarios incorporating differentiable scenario perturbations and compliance regularizers.Through these synergistic components, our approach delivers a unified solution for forecasting, anomaly detection, anddecision optimization in high-dimensional financial environments. Experimental results on simulated and real-world datasetsdemonstrate superior accuracy, compliance fidelity, and temporal stability, thus validating the utility of our method forpolicy-consistent risk prediction. This work contributes to the field by advancing interpretable, regulation-aware machinelearning frameworks capable of navigating the evolving landscape of financial technologies.","manuscriptTitle":"Multimodal deep learning framework for shadowbanking risk prediction - dynamic decisionoptimization integrating knowledge graph andreinforcement learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-14 05:42:44","doi":"10.21203/rs.3.rs-7351508/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-23T05:10:10+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-22T09:13:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-03T09:30:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"120545203348994366394578384530027531767","date":"2025-10-03T09:25:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"337640600576212659614315681801525119334","date":"2025-10-03T01:45:50+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-01T08:57:08+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-04T16:30:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-13T05:38:20+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-12T12:17:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-08-12T05:10:51+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3c602a43-219a-49de-974c-29b87c1fff9b","owner":[],"postedDate":"October 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":56216267,"name":"Physical sciences/Engineering"},{"id":56216268,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-03-18T15:23:39+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-14 05:42:44","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7351508","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7351508","identity":"rs-7351508","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}