| fullrecord |
[{"key": "dc.contributor.advisor", "value": "Haghparast, Majid", "language": null, "element": "contributor", "qualifier": "advisor", "schema": "dc"}, {"key": "dc.contributor.author", "value": "Sarker, Shaswato", "language": null, "element": "contributor", "qualifier": "author", "schema": "dc"}, {"key": "dc.date.accessioned", "value": "2025-06-02T11:56:47Z", "language": null, "element": "date", "qualifier": "accessioned", "schema": "dc"}, {"key": "dc.date.available", "value": "2025-06-02T11:56:47Z", "language": null, "element": "date", "qualifier": "available", "schema": "dc"}, {"key": "dc.date.issued", "value": "2025", "language": null, "element": "date", "qualifier": "issued", "schema": "dc"}, {"key": "dc.identifier.uri", "value": "https://jyx.jyu.fi/handle/123456789/102961", "language": null, "element": "identifier", "qualifier": "uri", "schema": "dc"}, {"key": "dc.description.abstract", "value": "Quantum computers accessed through the cloud may soon allow multiple users to share hardware simultaneously, easing resource constraints. That convenience creates the risk of quantum crosstalk, where operations on one qubit can perturb a neighboring one, serving as an attack vector. This thesis investigated whether classical machine-learning models, trained on simulated quantum data, can detect and quantify the degradation caused by crosstalk with no prior quantum knowledge. A total of 4566 ten-qubit MNISQ circuits encoding MNIST images were run on a simulator modeled after IBM\u2019s 127-qubit Brisbane device, with injected gate, relaxation, readout, and both coherent and stochastic crosstalk errors as noise models. From the resulting density matrices and original circuits, we built a dataset combining complex quantum features with simple classical descriptors like circuit depth and gate counts. We then designed TripleBranchNet, a model that merges three branches: a convolutional neural network (CNN) over the off-diagonal real and imaginary parts of the density matrix, a multilayer perceptron (MLP) for the diagonal terms, and another MLP for classical features. Trained on 80\\% of the data, the best model reached a mean absolute error of 0.0055 and an \\(R^{2}\\) score of 0.5131 on the unseen test set. The results show that crosstalk leaves a detectable imprint in the density matrix, one visible and learnable to classical AI models. This points to the possibility of software-level monitoring without hardware changes. Still, about 49\\% of variance remains unexplained, and real-device validation is required before deployment. The work offers a preliminary, practical step toward AI-assisted safeguards as quantum multi-tenancy grows with the proof-of-concept TripleBranchNet model.", "language": "en", "element": "description", "qualifier": "abstract", "schema": "dc"}, {"key": "dc.description.abstract", "value": "Pilvipalvelun kautta k\u00e4ytett\u00e4v\u00e4t kvanttitietokoneet voivat pian mahdollistaa useiden k\u00e4ytt\u00e4jien samanaikaisen laitteiston jakamisen, mik\u00e4 lievitt\u00e4\u00e4 resurssirajoitteita. T\u00e4m\u00e4 mukavuus lis\u00e4\u00e4 kuitenkin kvanttikytkeytymisen (crosstalk) riski\u00e4, jossa yhden kubitin operaatiot h\u00e4iritsev\u00e4t viereist\u00e4 kubittia ja voivat toimia hy\u00f6kk\u00e4ysvektorina. T\u00e4ss\u00e4 tutkielmassa selvitettiin, voivatko klassiset koneoppimismallit, jotka on koulutettu simuloidulla kvanttidatalla ilman aiempaa kvanttitietoa, havaita ja m\u00e4\u00e4r\u00e4llist\u00e4\u00e4 crosstalkin aiheuttaman heikkenemisen. Yhteens\u00e4 4566 kymmenkubittista MNISQ-piiri\u00e4, jotka koodasivat MNIST-kuvia, ajettiin IBM:n 127-kubittista Brisbane-laitetta j\u00e4ljittelev\u00e4ss\u00e4 simulaattorissa, johon injektoitiin portti-, relaksaatio- ja lukukohinaa sek\u00e4 koherentteja ja stokastisia crosstalk-virheit\u00e4. Tuloksena saaduista tiheysmatriiseista ja alkuper\u00e4isist\u00e4 piireist\u00e4 koottiin aineisto, jossa monimutkaiset kvanttiominaisuudet yhdistyiv\u00e4t yksinkertaisiin klassisiin kuvaajiin, kuten piirisyvyyteen ja porttim\u00e4\u00e4riin. Seuraavaksi suunniteltiin TripleBranchNet, joka yhdist\u00e4\u00e4 kolme haaraa: konvoluutioneuroverkon (CNN) tiheysmatriisin sivudiagonaalisten reaalisten ja imaginaaristen osien analysointiin, monikerrosperceptronin (MLP) sen diagonaalialkioille sek\u00e4 toisen MLP:n klassisille piirteille. Kun malli koulutettiin 80\\%:lla datasta, paras versio saavutti tuntemattomalla testijoukolla keskim\u00e4\u00e4r\u00e4isen absoluuttisen virheen 0,0055 ja \\(R^{2}\\)-arvon 0,5131. Tulokset osoittavat, ett\u00e4 crosstalk j\u00e4tt\u00e4\u00e4 tiheysmatriisiin havaittavan j\u00e4ljen, jonka klassiset teko\u00e4lymallit voivat oppia. T\u00e4m\u00e4 viittaa mahdollisuuteen ohjelmistopohjaiseen monitorointiin ilman laitteistomuutoksia. Noin 49\\% varianssista j\u00e4i kuitenkin selitt\u00e4m\u00e4tt\u00e4, ja todellisiin laitteisiin perustuva validointi on tarpeen ennen k\u00e4ytt\u00f6\u00f6nottoa. Ty\u00f6 tarjoaa alustavan, k\u00e4yt\u00e4nn\u00f6llisen askeleen kohti teko\u00e4lyavusteisia suojausmekanismeja kvanttimonik\u00e4ytt\u00f6isyyden laajentuessa, esitellen todisteena TripleBranchNet-mallin.", "language": "fi", "element": "description", "qualifier": "abstract", "schema": "dc"}, {"key": "dc.description.provenance", "value": "Submitted by jyx lomake-julkaisija (jyx-julkaisija.group@korppi.jyu.fi) on 2025-06-02T11:56:47Z\nNo. of bitstreams: 0", "language": "en", "element": "description", "qualifier": "provenance", "schema": "dc"}, {"key": "dc.description.provenance", "value": "Made available in DSpace on 2025-06-02T11:56:47Z (GMT). No. of bitstreams: 0", "language": "en", "element": "description", "qualifier": "provenance", "schema": "dc"}, {"key": "dc.format.extent", "value": "166", "language": null, "element": "format", "qualifier": "extent", "schema": "dc"}, {"key": "dc.format.mimetype", "value": "application/pdf", "language": null, "element": "format", "qualifier": "mimetype", "schema": "dc"}, {"key": "dc.language.iso", "value": "eng", "language": null, "element": "language", "qualifier": "iso", "schema": "dc"}, {"key": "dc.rights", "value": "CC BY 4.0", "language": null, "element": "rights", "qualifier": null, "schema": "dc"}, {"key": "dc.title", "value": "Toward Secure Multi-Tenancy in Quantum Computing: A Software-First Approach to Detecting Crosstalk-Induced Fidelity Degradation with Machine Learning", "language": null, "element": "title", "qualifier": null, "schema": "dc"}, {"key": "dc.type", "value": "master thesis", "language": null, "element": "type", "qualifier": null, "schema": "dc"}, {"key": "dc.identifier.urn", "value": "URN:NBN:fi:jyu-202506024770", "language": null, "element": "identifier", "qualifier": "urn", "schema": "dc"}, {"key": "dc.contributor.faculty", "value": "Informaatioteknologian tiedekunta", "language": "fi", "element": "contributor", "qualifier": "faculty", "schema": "dc"}, {"key": "dc.contributor.faculty", "value": "Faculty of Information Technology", "language": "en", "element": "contributor", "qualifier": "faculty", "schema": "dc"}, {"key": "dc.contributor.organization", "value": "Jyv\u00e4skyl\u00e4n yliopisto", "language": "fi", "element": "contributor", "qualifier": "organization", "schema": "dc"}, {"key": "dc.contributor.organization", "value": "University of Jyv\u00e4skyl\u00e4", "language": "en", "element": "contributor", "qualifier": "organization", "schema": "dc"}, {"key": "dc.subject.discipline", "value": "Master's Degree Programme in Artificial Intelligence", "language": "fi", "element": "subject", "qualifier": "discipline", "schema": "dc"}, {"key": "dc.subject.discipline", "value": "Master's Degree Programme in Artificial Intelligence", "language": "en", "element": "subject", "qualifier": "discipline", "schema": "dc"}, {"key": "dc.type.coar", "value": "http://purl.org/coar/resource_type/c_bdcc", "language": null, "element": "type", "qualifier": "coar", "schema": "dc"}, {"key": "dc.rights.copyright", "value": "\u00a9 The Author(s)", "language": null, "element": "rights", "qualifier": "copyright", "schema": "dc"}, {"key": "dc.rights.accesslevel", "value": "openAccess", "language": null, "element": "rights", "qualifier": "accesslevel", "schema": "dc"}, {"key": "dc.type.publication", "value": "masterThesis", "language": null, "element": "type", "qualifier": "publication", "schema": "dc"}, {"key": "dc.format.content", "value": "fulltext", "language": null, "element": "format", "qualifier": "content", "schema": "dc"}, {"key": "dc.rights.url", "value": "https://creativecommons.org/licenses/by/4.0/", "language": null, "element": "rights", "qualifier": "url", "schema": "dc"}, {"key": "dc.description.accessibilityfeature", "value": "ei tietoa saavutettavuudesta", "language": "fi", "element": "description", "qualifier": "accessibilityfeature", "schema": "dc"}, {"key": "dc.description.accessibilityfeature", "value": "unknown accessibility", "language": "en", "element": "description", "qualifier": "accessibilityfeature", "schema": "dc"}]
|