{"id":{"repo_id":"utahstate","oai_identifier":"oai:digitalcommons.usu.edu:etd-1535"},"canonical_url":"https://search.ndltd.org/etd/utahstate/oai:digitalcommons.usu.edu:etd-1535","repository":{"repo_id":"utahstate","name":"Utah State University","base_url":"https://digitalcommons.usu.edu/do/oai/"},"display":{"title":"Neural Networks and the Natural Gradient","abstract":"Neural network training algorithms have always suffered from the problem of local minima. The advent of natural gradient algorithms promised to overcome this shortcoming by finding better local minima. However, they require additional training parameters and computational overhead. By using a new formulation for the natural gradient, an algorithm is described that uses less memory and processing time than previous algorithms with comparable performance.","abstract_html":"Neural network training algorithms have always suffered from the problem of local minima. The advent of natural gradient algorithms promised to overcome this shortcoming by finding better local minima. However, they require additional training parameters and computational overhead. By using a new formulation for the natural gradient, an algorithm is described that uses less memory and processing time than previous algorithms with comparable performance.","abstract_has_math":false,"creators":["Bastian, Michael R."],"institution":"Hosted by Utah State University Libraries","degree_name":"Doctor of Philosophy (PhD)","degree_level":"Dissertation","degree_discipline":"Electrical and Computer Engineering","degree_department":null,"school":null,"contributors":["Jacob H. Gunther","Todd K. Moon","YangQuan Chen"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:46:18Z","subjects":["Backpropagation","Fisher Information Matrix","Natural Gradient","Neural Networks","Newton Methods","Riemannian Geometry","Electrical and Computer Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.26076/48b1-9bb8"],"render_values":[{"text":"https://doi.org/10.26076/48b1-9bb8","href":"https://doi.org/10.26076/48b1-9bb8","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.usu.edu/etd/539","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://digitalcommons.usu.edu/etd/539","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://digitalcommons.usu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.usu.edu%3Aetd-1535","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Jacob H. Gunther","Todd K. Moon","YangQuan Chen"]},{"key":"dc:creator","label":"Author","values":["Bastian, Michael R."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2010-02-09T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2010-05-01T07:00:00Z"]},{"key":"dc:publisher","label":"Institution","values":["Hosted by Utah State University Libraries"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical and Computer Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy (PhD)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Backpropagation","Fisher Information Matrix","Natural Gradient","Neural Networks","Newton Methods","Riemannian Geometry","Electrical and Computer Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.usu.edu/etd/539"]},{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.26076/48b1-9bb8"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Neural network training algorithms have always suffered from the problem of local minima. The advent of natural gradient algorithms promised to overcome this shortcoming by finding better local minima. However, they require additional training parameters and computational overhead. By using a new formulation for the natural gradient, an algorithm is described that uses less memory and processing time than previous algorithms with comparable performance."]},{"key":"dc:title","label":"Title","values":["Neural Networks and the Natural Gradient"]}]}],"canonical_facts":{"dc:contributor":["Jacob H. Gunther","Todd K. Moon","YangQuan Chen"],"dc:creator":["Bastian, Michael R."],"dc:date.available":["2010-02-09T08:00:00Z"],"dc:date.created":["2010-05-01T07:00:00Z"],"dc:description.abstract":["Neural network training algorithms have always suffered from the problem of local minima. The advent of natural gradient algorithms promised to overcome this shortcoming by finding better local minima. However, they require additional training parameters and computational overhead. By using a new formulation for the natural gradient, an algorithm is described that uses less memory and processing time than previous algorithms with comparable performance."],"dc:identifier":["https://digitalcommons.usu.edu/etd/539"],"dc:identifier.doi":["https://doi.org/10.26076/48b1-9bb8"],"dc:publisher":["Hosted by Utah State University Libraries"],"dc:subject":["Backpropagation","Fisher Information Matrix","Natural Gradient","Neural Networks","Newton Methods","Riemannian Geometry","Electrical and Computer Engineering"],"dc:title":["Neural Networks and the Natural Gradient"],"thesis:degree_discipline":["Electrical and Computer Engineering"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Doctor of Philosophy (PhD)"]},"updated_at":"2026-11-19T08:46:18Z"}