{"id":{"repo_id":"etsu","oai_identifier":"oai:dc.etsu.edu:etd-6160"},"canonical_url":"https://search.ndltd.org/etd/etsu/oai:dc.etsu.edu:etd-6160","repository":{"repo_id":"etsu","name":"East Tennessee State University","base_url":"https://dc.etsu.edu/do/oai/"},"display":{"title":"Grokking Applied to Chaotic Iterates of the Logistic Map","abstract":"This thesis investigates grokking, the delayed transition from memorization to generalization in neural networks trained on deterministic chaotic data. Using an integer–arithmetic discretization of the logistic map, yn+1 =( a yn(p − yn))/ p 2 , bounded aperiodic sequences were generated across control parameters α ranging from 3.0 to 4.0. Transformer-based models displayed characteristic grokking curves. In periodic and chaotic regimes, validation accuracy rose suddenly after long plateaus, while at the Feigenbaum boundary (α ≈ 3.57) generalization failed completely. Increasing data diversity restored learning in chaotic domains, and explicit α–conditioning enabled a single network to generalize across all regimes. A bifurcation diagram of model-predicted data reproduced the main features of the true logistic map, confirming that the network captured the underlying dynamics. These results link delayed generalization in deep learning to structural transitions in deterministic chaos.","abstract_html":"This thesis investigates grokking, the delayed transition from memorization to generalization in neural networks trained on deterministic chaotic data. Using an integer–arithmetic discretization of the logistic map, yn+1 =( a yn(p − yn))/ p 2 , bounded aperiodic sequences were generated across control parameters α ranging from 3.0 to 4.0. Transformer-based models displayed characteristic grokking curves. In periodic and chaotic regimes, validation accuracy rose suddenly after long plateaus, while at the Feigenbaum boundary (α ≈ 3.57) generalization failed completely. Increasing data diversity restored learning in chaotic domains, and explicit α–conditioning enabled a single network to generalize across all regimes. A bifurcation diagram of model-predicted data reproduced the main features of the true logistic map, confirming that the network captured the underlying dynamics. These results link delayed generalization in deep learning to structural transitions in deterministic chaos.","abstract_has_math":false,"creators":["Donkoh, Felix"],"institution":"East Tennessee State University","degree_name":"MS (Master of Science)","degree_level":"Thesis - unrestricted","degree_discipline":"Mathematical Sciences","degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-12-01T08:00:00Z","date_published":"2025-12-01T08:00:00Z","updated_at":"2026-11-19T07:57:23Z","subjects":["grokking","neural network","logistic map","chaos","bifurcation.","Data Science","Non-linear Dynamics","Physical Sciences and Mathematics"],"languages":["eng"],"rights":["Copyright by the authors."],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://dc.etsu.edu/etd/4613","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://dc.etsu.edu/etd/4613","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":"https://dc.etsu.edu/context/etd/article/6160/viewcontent/DonkohF112025f.pdf","download_label":"Download PDF","download_source":"dc:identifier","access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://dc.etsu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adc.etsu.edu%3Aetd-6160","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Donkoh, Felix"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2025-11-05T08:00:00Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-12-01T08:00:00Z"]},{"key":"dc:publisher","label":"Institution","values":["East Tennessee State University"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Mathematical Sciences"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis - unrestricted"]},{"key":"thesis:degree_name","label":"Degree Name","values":["MS (Master of Science)"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["East Tennessee State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["grokking","neural network","logistic map","chaos","bifurcation.","Data Science","Non-linear Dynamics","Physical Sciences and Mathematics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright by the authors."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://dc.etsu.edu/context/etd/article/6160/viewcontent/DonkohF112025f.pdf","https://dc.etsu.edu/etd/4613"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis investigates grokking, the delayed transition from memorization to generalization in neural networks trained on deterministic chaotic data. Using an integer–arithmetic discretization of the logistic map, yn+1 =( a yn(p − yn))/ p 2 , bounded aperiodic sequences were generated across control parameters α ranging from 3.0 to 4.0. Transformer-based models displayed characteristic grokking curves. In periodic and chaotic regimes, validation accuracy rose suddenly after long plateaus, while at the Feigenbaum boundary (α ≈ 3.57) generalization failed completely. Increasing data diversity restored learning in chaotic domains, and explicit α–conditioning enabled a single network to generalize across all regimes. A bifurcation diagram of model-predicted data reproduced the main features of the true logistic map, confirming that the network captured the underlying dynamics. These results link delayed generalization in deep learning to structural transitions in deterministic chaos."]},{"key":"dc:title","label":"Title","values":["Grokking Applied to Chaotic Iterates of the Logistic Map"]}]}],"canonical_facts":{"dc:creator":["Donkoh, Felix"],"dc:date.available":["2025-11-05T08:00:00Z"],"dc:date.issued":["2025-12-01T08:00:00Z"],"dc:description.abstract":["This thesis investigates grokking, the delayed transition from memorization to generalization in neural networks trained on deterministic chaotic data. Using an integer–arithmetic discretization of the logistic map, yn+1 =( a yn(p − yn))/ p 2 , bounded aperiodic sequences were generated across control parameters α ranging from 3.0 to 4.0. Transformer-based models displayed characteristic grokking curves. In periodic and chaotic regimes, validation accuracy rose suddenly after long plateaus, while at the Feigenbaum boundary (α ≈ 3.57) generalization failed completely. Increasing data diversity restored learning in chaotic domains, and explicit α–conditioning enabled a single network to generalize across all regimes. A bifurcation diagram of model-predicted data reproduced the main features of the true logistic map, confirming that the network captured the underlying dynamics. These results link delayed generalization in deep learning to structural transitions in deterministic chaos."],"dc:identifier":["https://dc.etsu.edu/context/etd/article/6160/viewcontent/DonkohF112025f.pdf","https://dc.etsu.edu/etd/4613"],"dc:language":["eng"],"dc:publisher":["East Tennessee State University"],"dc:rights":["Copyright by the authors."],"dc:subject":["grokking","neural network","logistic map","chaos","bifurcation.","Data Science","Non-linear Dynamics","Physical Sciences and Mathematics"],"dc:title":["Grokking Applied to Chaotic Iterates of the Logistic Map"],"thesis:degree_discipline":["Mathematical Sciences"],"thesis:degree_level":["Thesis - unrestricted"],"thesis:degree_name":["MS (Master of Science)"],"thesis:institution_name":["East Tennessee State University"]},"updated_at":"2026-11-19T07:57:23Z"}