{"id":{"repo_id":"texas","oai_identifier":"oai:repositories.lib.utexas.edu:2152/114781"},"canonical_url":"https://search.dev.ndltd.org/etd/texas/oai:repositories.lib.utexas.edu:2152/114781","repository":{"repo_id":"texas","name":"University of Texas","base_url":"https://repositories.lib.utexas.edu/server/oai/request"},"display":{"title":"Components and principles of streaming principal components","abstract":"Principal Component Analysis (PCA) is a fundamental pillar of modern data pipelines, but its traditional implementation is woefully inadequate for modern data problems. In this work we present our contributions to the field of streaming principal component analysis---research that adds critical flexibility to one of the most common tools in optimization. This includes a practical new algorithm--AdaOja--which we outline in the context of both streaming principal component analysis and streaming kernel principal component analysis. We also present new mathematical theory inspired by our study of theoretical convergence for this algorithm. Streaming principal component analysis and streaming kernel principal component analysis can be seamlessly integrated into many applications with significant improvements in scalability. We specifically demonstrate the considerable improvements that can be achieved by applying streaming KPCA to kernel analog forecasting (KAF) for dynamical systems.","abstract_html":"Principal Component Analysis (PCA) is a fundamental pillar of modern data pipelines, but its traditional implementation is woefully inadequate for modern data problems. In this work we present our contributions to the field of streaming principal component analysis---research that adds critical flexibility to one of the most common tools in optimization. This includes a practical new algorithm--AdaOja--which we outline in the context of both streaming principal component analysis and streaming kernel principal component analysis. We also present new mathematical theory inspired by our study of theoretical convergence for this algorithm. Streaming principal component analysis and streaming kernel principal component analysis can be seamlessly integrated into many applications with significant improvements in scalability. We specifically demonstrate the considerable improvements that can be achieved by applying streaming KPCA to kernel analog forecasting (KAF) for dynamical systems.","abstract_has_math":false,"creators":["Henriksen, Amelia"],"institution":"The University of Texas at Austin","degree_name":"Doctor of Philosophy","degree_level":"Doctoral","degree_discipline":"Computational Science, Engineering, and Mathematics","degree_department":null,"school":null,"contributors":[],"advisors":["Ward, Rachel, 1983-"],"committee_chairs":[],"committee_members":["Bui-Thanh, Tan","Martinsson, Per Gunnar","Price, Eric","Topcu, Ufuk"],"year":2021,"date_issued":"2021-08","date_published":"2021-08","updated_at":"2026-07-24T05:01:08Z","subjects":["Dimensionality reduction","Online algorithms","Kernel analog forecasting"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://dx.doi.org/10.26153/tsw/41684"],"render_values":[{"text":"http://dx.doi.org/10.26153/tsw/41684","href":"http://dx.doi.org/10.26153/tsw/41684","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/2152/114781","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Ward, Rachel, 1983-"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Bui-Thanh, Tan","Martinsson, Per Gunnar","Price, Eric","Topcu, Ufuk"]},{"key":"dc:creator","label":"Author","values":["Henriksen, Amelia"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2022-07-01T14:45:14Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2022-07-01T14:45:14Z"]},{"key":"dc:date.issued","label":"Date","values":["2021-08"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computational Science, Engineering, and Mathematics"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["The University of Texas at Austin"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Dimensionality reduction","Online algorithms","Kernel analog forecasting"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2152/114781","http://dx.doi.org/10.26153/tsw/41684"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Principal Component Analysis (PCA) is a fundamental pillar of modern data pipelines, but its traditional implementation is woefully inadequate for modern data problems. In this work we present our contributions to the field of streaming principal component analysis---research that adds critical flexibility to one of the most common tools in optimization. This includes a practical new algorithm--AdaOja--which we outline in the context of both streaming principal component analysis and streaming kernel principal component analysis. We also present new mathematical theory inspired by our study of theoretical convergence for this algorithm. Streaming principal component analysis and streaming kernel principal component analysis can be seamlessly integrated into many applications with significant improvements in scalability. We specifically demonstrate the considerable improvements that can be achieved by applying streaming KPCA to kernel analog forecasting (KAF) for dynamical systems."]},{"key":"dc:format.mimetype","label":"Dc Format Mimetype","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Components and principles of streaming principal components"]}]}],"canonical_facts":{"dc:contributor.advisor":["Ward, Rachel, 1983-"],"dc:contributor.committeemember":["Bui-Thanh, Tan","Martinsson, Per Gunnar","Price, Eric","Topcu, Ufuk"],"dc:creator":["Henriksen, Amelia"],"dc:date.accessioned":["2022-07-01T14:45:14Z"],"dc:date.available":["2022-07-01T14:45:14Z"],"dc:date.issued":["2021-08"],"dc:description.abstract":["Principal Component Analysis (PCA) is a fundamental pillar of modern data pipelines, but its traditional implementation is woefully inadequate for modern data problems. In this work we present our contributions to the field of streaming principal component analysis---research that adds critical flexibility to one of the most common tools in optimization. This includes a practical new algorithm--AdaOja--which we outline in the context of both streaming principal component analysis and streaming kernel principal component analysis. We also present new mathematical theory inspired by our study of theoretical convergence for this algorithm. Streaming principal component analysis and streaming kernel principal component analysis can be seamlessly integrated into many applications with significant improvements in scalability. We specifically demonstrate the considerable improvements that can be achieved by applying streaming KPCA to kernel analog forecasting (KAF) for dynamical systems."],"dc:format.mimetype":["application/pdf"],"dc:identifier.uri":["https://hdl.handle.net/2152/114781","http://dx.doi.org/10.26153/tsw/41684"],"dc:language.iso":["en"],"dc:subject":["Dimensionality reduction","Online algorithms","Kernel analog forecasting"],"dc:title":["Components and principles of streaming principal components"],"dc:type":["Thesis"],"thesis:degree_discipline":["Computational Science, Engineering, and Mathematics"],"thesis:degree_level":["Doctoral"],"thesis:degree_name":["Doctor of Philosophy"],"thesis:institution_name":["The University of Texas at Austin"]},"updated_at":"2026-07-24T05:01:08Z"}