{"id":{"repo_id":"rit","oai_identifier":"oai:repository.rit.edu:theses-13232"},"canonical_url":"https://search.ndltd.org/etd/rit/oai:repository.rit.edu:theses-13232","repository":{"repo_id":"rit","name":"Rochester Institute of Technology","base_url":"https://repository.rit.edu/do/oai/"},"display":{"title":"Visualizing the Dynamics of Neuroevolution with Genetic Distance Projections","abstract":"Evolutionary algorithms have shown substantial progress in recent years, especially in neural architecture search and neuroevolution applications. Despite their effectiveness, analyzing and understanding the evolutionary paths these algorithms traverse to reach solutions remains challenging. These algorithms often involve distributed computing strategies, which can include subpopulations or islands, and they explore massive or even unbounded search spaces in both continuous and non-continuous domains. Manually examining individual solutions to understand the evolutionary dynamics is often infeasible due to large population sizes, large genome sizes, and high generation counts. This work introduces a new methodology for visualizing neuroevolution population dynamics called genetic distance projections, along with a novel neural network-based method for generating these representations. This methodology is evaluated empirically and is found to perform better than other, more traditional methods in generating these representations. The usefulness of this methodology was validated with three different neuroevolution frameworks, including NEAT, EXAMM, and an experimental one, EXA-STAR. In addition, the debugging potential of this methodology is demonstrated on EXAMM.","abstract_html":"Evolutionary algorithms have shown substantial progress in recent years, especially in neural architecture search and neuroevolution applications. Despite their effectiveness, analyzing and understanding the evolutionary paths these algorithms traverse to reach solutions remains challenging. These algorithms often involve distributed computing strategies, which can include subpopulations or islands, and they explore massive or even unbounded search spaces in both continuous and non-continuous domains. Manually examining individual solutions to understand the evolutionary dynamics is often infeasible due to large population sizes, large genome sizes, and high generation counts. This work introduces a new methodology for visualizing neuroevolution population dynamics called genetic distance projections, along with a novel neural network-based method for generating these representations. This methodology is evaluated empirically and is found to perform better than other, more traditional methods in generating these representations. The usefulness of this methodology was validated with three different neuroevolution frameworks, including NEAT, EXAMM, and an experimental one, EXA-STAR. In addition, the debugging potential of this methodology is demonstrated on EXAMM.","abstract_has_math":false,"creators":["Patterson, Evan"],"institution":null,"degree_name":"Artificial Intelligence (MS)","degree_level":"Thesis","degree_discipline":"Information, School of","degree_department":null,"school":null,"contributors":["Travis Desell"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T09:05:51Z","subjects":["Evolutionary algorithms","Genetic algorithms","Neuroevolution"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://repository.rit.edu/theses/12094","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://repository.rit.edu/theses/12094","record_label":"Repository record","record_kind":"record","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://repository.rit.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Arepository.rit.edu%3Atheses-13232","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Travis Desell"]},{"key":"dc:creator","label":"Author","values":["Patterson, Evan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2025-05-16T07:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2025-05-02T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Information, School of"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Artificial Intelligence (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Evolutionary algorithms","Genetic algorithms","Neuroevolution"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://repository.rit.edu/theses/12094"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Evolutionary algorithms have shown substantial progress in recent years, especially in neural architecture search and neuroevolution applications. Despite their effectiveness, analyzing and understanding the evolutionary paths these algorithms traverse to reach solutions remains challenging. These algorithms often involve distributed computing strategies, which can include subpopulations or islands, and they explore massive or even unbounded search spaces in both continuous and non-continuous domains. Manually examining individual solutions to understand the evolutionary dynamics is often infeasible due to large population sizes, large genome sizes, and high generation counts. This work introduces a new methodology for visualizing neuroevolution population dynamics called genetic distance projections, along with a novel neural network-based method for generating these representations. This methodology is evaluated empirically and is found to perform better than other, more traditional methods in generating these representations. The usefulness of this methodology was validated with three different neuroevolution frameworks, including NEAT, EXAMM, and an experimental one, EXA-STAR. In addition, the debugging potential of this methodology is demonstrated on EXAMM."]},{"key":"dc:title","label":"Title","values":["Visualizing the Dynamics of Neuroevolution with Genetic Distance Projections"]}]}],"canonical_facts":{"dc:contributor":["Travis Desell"],"dc:creator":["Patterson, Evan"],"dc:date.available":["2025-05-16T07:00:00Z"],"dc:date.created":["2025-05-02T07:00:00Z"],"dc:description.abstract":["Evolutionary algorithms have shown substantial progress in recent years, especially in neural architecture search and neuroevolution applications. Despite their effectiveness, analyzing and understanding the evolutionary paths these algorithms traverse to reach solutions remains challenging. These algorithms often involve distributed computing strategies, which can include subpopulations or islands, and they explore massive or even unbounded search spaces in both continuous and non-continuous domains. Manually examining individual solutions to understand the evolutionary dynamics is often infeasible due to large population sizes, large genome sizes, and high generation counts. This work introduces a new methodology for visualizing neuroevolution population dynamics called genetic distance projections, along with a novel neural network-based method for generating these representations. This methodology is evaluated empirically and is found to perform better than other, more traditional methods in generating these representations. The usefulness of this methodology was validated with three different neuroevolution frameworks, including NEAT, EXAMM, and an experimental one, EXA-STAR. In addition, the debugging potential of this methodology is demonstrated on EXAMM."],"dc:identifier":["https://repository.rit.edu/theses/12094"],"dc:subject":["Evolutionary algorithms","Genetic algorithms","Neuroevolution"],"dc:title":["Visualizing the Dynamics of Neuroevolution with Genetic Distance Projections"],"thesis:degree_discipline":["Information, School of"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Artificial Intelligence (MS)"]},"updated_at":"2026-11-19T09:05:51Z"}