{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_52934"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_52934","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"Transitioning from Multi- to Massive-Objective Optimization via Theoretically-Informed Algorithm Selection with Applications in Fair Machine Learning","abstract":"Thesis (Ph.D.)--Michigan State University. Computer Science - Doctor of Philosophy, 2026","abstract_html":"Thesis (Ph.D.)--Michigan State University. Computer Science - Doctor of Philosophy, 2026","abstract_has_math":false,"creators":["Shahbandegan, Shakiba"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Dolson, Emily"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-18T18:59:31Z","subjects":["Computer science"],"languages":["eng"],"rights":["Attribution 4.0 International"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52934","local:Shahbandegan_grad.msu_0128D_21939"],"render_values":[{"text":"etd:52934","href":null,"code":true},{"text":"local:Shahbandegan_grad.msu_0128D_21939","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/sv7j-1620","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/sv7j-1620","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_52934","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dolson, Emily"]},{"key":"dc:creator","label":"Author","values":["Shahbandegan, Shakiba"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["text","Electronic dissertations"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computer science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52934","local:Shahbandegan_grad.msu_0128D_21939","https://doi.org/doi:10.25335/sv7j-1620"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (Ph.D.)--Michigan State University. Computer Science - Doctor of Philosophy, 2026","Multi-objective optimization is an important research topic that concerns problems involving multiple objectives that must be optimized simultaneously. When the number of objectives exceeds three, the problem enters the domain of many-objective optimization, where new challenges emerge due to the increased complexity and high computational cost of high-dimensional objective spaces. This dissertation investigates these problems through both theoretical and empirical analysis. First, the performance of lexicase selection is examined in a maximally conflicting optimization problem with up to 200 objectives. Results show that even though lexicase selection was originally developed for genetic programming, it demonstrates competitive performance in multi- and many objective optimization problems provided that the population size scales appropriately with dimensionality. Next, the dissertation explores two fundamental yet under-studied questions: How do objectives interact with one another? Are they always conflicting, as commonly assumed, or can they exhibit neutral or even synergistic relationships and how does this affect algorithm performance? Furthermore, what happens when the number of objectives grows far beyond the typical experimental settings and becomes a massive-objective optimization problem? We examine the performance of state-of the-art evolutionary algorithms under different objective interactions and dimensionalities and show that understanding the relationships between the objectives is essential for effective algorithm selection. Finally, a practical case study on evolving fair classifiers is presented in which the number of objectives can become massive depending on the dataset and the formulation of the problem. This study also indicates how subgroup definitions influence both the accuracy and fairness of machine learning models. Together, these findings advance our understanding of how objective interactions and dimensionality shape algorithmic dynamics and establish lexicase selection as a robust approach for multi-, many-, and massive-objective optimization.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf","1 online resource (125 p.) : ill."]},{"key":"dc:title","label":"Title","values":["Transitioning from Multi- to Massive-Objective Optimization via Theoretically-Informed Algorithm Selection with Applications in Fair Machine Learning"]}]}],"canonical_facts":{"dc:contributor":["Dolson, Emily"],"dc:creator":["Shahbandegan, Shakiba"],"dc:date":["2026"],"dc:description":["Thesis (Ph.D.)--Michigan State University. Computer Science - Doctor of Philosophy, 2026","Multi-objective optimization is an important research topic that concerns problems involving multiple objectives that must be optimized simultaneously. When the number of objectives exceeds three, the problem enters the domain of many-objective optimization, where new challenges emerge due to the increased complexity and high computational cost of high-dimensional objective spaces. This dissertation investigates these problems through both theoretical and empirical analysis. First, the performance of lexicase selection is examined in a maximally conflicting optimization problem with up to 200 objectives. Results show that even though lexicase selection was originally developed for genetic programming, it demonstrates competitive performance in multi- and many objective optimization problems provided that the population size scales appropriately with dimensionality. Next, the dissertation explores two fundamental yet under-studied questions: How do objectives interact with one another? Are they always conflicting, as commonly assumed, or can they exhibit neutral or even synergistic relationships and how does this affect algorithm performance? Furthermore, what happens when the number of objectives grows far beyond the typical experimental settings and becomes a massive-objective optimization problem? We examine the performance of state-of the-art evolutionary algorithms under different objective interactions and dimensionalities and show that understanding the relationships between the objectives is essential for effective algorithm selection. Finally, a practical case study on evolving fair classifiers is presented in which the number of objectives can become massive depending on the dataset and the formulation of the problem. This study also indicates how subgroup definitions influence both the accuracy and fairness of machine learning models. Together, these findings advance our understanding of how objective interactions and dimensionality shape algorithmic dynamics and establish lexicase selection as a robust approach for multi-, many-, and massive-objective optimization.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."],"dc:format":["application/pdf","1 online resource (125 p.) : ill."],"dc:identifier":["etd:52934","local:Shahbandegan_grad.msu_0128D_21939","https://doi.org/doi:10.25335/sv7j-1620"],"dc:language":["eng"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["Attribution 4.0 International"],"dc:subject":["Computer science"],"dc:title":["Transitioning from Multi- to Massive-Objective Optimization via Theoretically-Informed Algorithm Selection with Applications in Fair Machine Learning"],"dc:type":["text","Electronic dissertations"]},"updated_at":"2026-11-18T18:59:31Z"}