{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/110594"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/110594","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Unsupervised anomaly detection in multi-class datasets using Generative Adversarial Networks","abstract":"\"Presented in this thesis is a novel Generative Adversarial Network, or GAN, based method, D-AnoGAN, for detecting anomalies in complex datasets containing disconnected data manifolds. Current state-of-the-art methods treat disconnected data manifolds as a single, continuous one to learn from. The key contribution of D-AnoGAN is specifically accounting for the discontinuity between manifolds within a dataset during training. To achieve this, a multi-generator network is first implemented, where each generator is responsible for learning a unique manifold of data. Second, a machine learning mechanism called a ''bandit\"\" is implemented to find the optimal set of generators required to cover all data manifolds through unsupervised prior-learning. Finally, the multi-generator and bandit are used to cluster data from the same manifold together during training, allowing them to be learned in a disconnected fashion. The proposed method's effectiveness is demonstrated on two publicly available datasets, as well as a new experimental dataset developed in-house, where state-of-the-art results are achieved.\"","abstract_html":""Presented in this thesis is a novel Generative Adversarial Network, or GAN, based method, D-AnoGAN, for detecting anomalies in complex datasets containing disconnected data manifolds. Current state-of-the-art methods treat disconnected data manifolds as a single, continuous one to learn from. The key contribution of D-AnoGAN is specifically accounting for the discontinuity between manifolds within a dataset during training. To achieve this, a multi-generator network is first implemented, where each generator is responsible for learning a unique manifold of data. Second, a machine learning mechanism called a ''bandit"" is implemented to find the optimal set of generators required to cover all data manifolds through unsupervised prior-learning. Finally, the multi-generator and bandit are used to cluster data from the same manifold together during training, allowing them to be learned in a disconnected fashion. The proposed method's effectiveness is demonstrated on two publicly available datasets, as well as a new experimental dataset developed in-house, where state-of-the-art results are achieved."","abstract_has_math":false,"creators":["Dimon, Walker Lee"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Aerospace Engineering","degree_department":null,"school":null,"contributors":["Lembeck, Michael F","Tran, Huy T"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021-09-17T01:13:33Z","date_published":"2021-09-17T01:13:33Z","updated_at":"2026-07-22T22:24:52Z","subjects":["Generative Adversarial Networks, Unsupervised Anomaly Detection, Computer Vision, Unsupervised Learning, Machine Learning"],"languages":["en"],"rights":["Copyright 2021 Walker Dimon"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/110594","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Lembeck, Michael F","Tran, Huy T"]},{"key":"dc:creator","label":"Author","values":["Dimon, Walker Lee"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2021-09-17T01:13:33Z","2021-04-28","2021-05"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Aerospace Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Generative Adversarial Networks, Unsupervised Anomaly Detection, Computer Vision, Unsupervised Learning, Machine Learning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2021 Walker Dimon"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/110594"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["\"Presented in this thesis is a novel Generative Adversarial Network, or GAN, based method, D-AnoGAN, for detecting anomalies in complex datasets containing disconnected data manifolds. Current state-of-the-art methods treat disconnected data manifolds as a single, continuous one to learn from. The key contribution of D-AnoGAN is specifically accounting for the discontinuity between manifolds within a dataset during training. To achieve this, a multi-generator network is first implemented, where each generator is responsible for learning a unique manifold of data. Second, a machine learning mechanism called a ''bandit\"\" is implemented to find the optimal set of generators required to cover all data manifolds through unsupervised prior-learning. Finally, the multi-generator and bandit are used to cluster data from the same manifold together during training, allowing them to be learned in a disconnected fashion. The proposed method's effectiveness is demonstrated on two publicly available datasets, as well as a new experimental dataset developed in-house, where state-of-the-art results are achieved.\"","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2021-09-16 without embargo terms","The student, Walker Dimon, accepted the attached license on 2021-04-28 at 12:00.","The student, Walker Dimon, submitted this Thesis for approval on 2021-04-28 at 12:11.","This Thesis was approved for publication on 2021-04-28 at 14:59.","DSpace SAF Submission Ingestion Package generated from Vireo submission #16602 on 2021-09-16 at 16:49:20","Made available in DSpace on 2021-09-17T01:13:33Z (GMT). No. of bitstreams: 3 DIMON-THESIS-2021.pdf: 3148125 bytes, checksum: d3bc6e7f38d2f2ad7a39fa5522797d3d (MD5) DIMON-THESIS-LATEX-FILES.zip: 3084895 bytes, checksum: 66a83e5b7dfe9045f24fa040e9eb6c17 (MD5) LICENSE.txt: 4209 bytes, checksum: 991245f6df61b80c605d4fcc37ae04c0 (MD5) Previous issue date: 2021-04-28"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Unsupervised anomaly detection in multi-class datasets using Generative Adversarial Networks"]}]}],"canonical_facts":{"dc:contributor":["Lembeck, Michael F","Tran, Huy T"],"dc:creator":["Dimon, Walker Lee"],"dc:date":["2021-09-17T01:13:33Z","2021-04-28","2021-05"],"dc:description":["\"Presented in this thesis is a novel Generative Adversarial Network, or GAN, based method, D-AnoGAN, for detecting anomalies in complex datasets containing disconnected data manifolds. Current state-of-the-art methods treat disconnected data manifolds as a single, continuous one to learn from. The key contribution of D-AnoGAN is specifically accounting for the discontinuity between manifolds within a dataset during training. To achieve this, a multi-generator network is first implemented, where each generator is responsible for learning a unique manifold of data. Second, a machine learning mechanism called a ''bandit\"\" is implemented to find the optimal set of generators required to cover all data manifolds through unsupervised prior-learning. Finally, the multi-generator and bandit are used to cluster data from the same manifold together during training, allowing them to be learned in a disconnected fashion. The proposed method's effectiveness is demonstrated on two publicly available datasets, as well as a new experimental dataset developed in-house, where state-of-the-art results are achieved.\"","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2021-09-16 without embargo terms","The student, Walker Dimon, accepted the attached license on 2021-04-28 at 12:00.","The student, Walker Dimon, submitted this Thesis for approval on 2021-04-28 at 12:11.","This Thesis was approved for publication on 2021-04-28 at 14:59.","DSpace SAF Submission Ingestion Package generated from Vireo submission #16602 on 2021-09-16 at 16:49:20","Made available in DSpace on 2021-09-17T01:13:33Z (GMT). No. of bitstreams: 3 DIMON-THESIS-2021.pdf: 3148125 bytes, checksum: d3bc6e7f38d2f2ad7a39fa5522797d3d (MD5) DIMON-THESIS-LATEX-FILES.zip: 3084895 bytes, checksum: 66a83e5b7dfe9045f24fa040e9eb6c17 (MD5) LICENSE.txt: 4209 bytes, checksum: 991245f6df61b80c605d4fcc37ae04c0 (MD5) Previous issue date: 2021-04-28"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/110594"],"dc:language":["en"],"dc:rights":["Copyright 2021 Walker Dimon"],"dc:subject":["Generative Adversarial Networks, Unsupervised Anomaly Detection, Computer Vision, Unsupervised Learning, Machine Learning"],"dc:title":["Unsupervised anomaly detection in multi-class datasets using Generative Adversarial Networks"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Aerospace Engineering"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:24:52Z"}