{"id":"https://openalex.org/W4416366417","doi":"https://doi.org/10.1109/tpami.2025.3634391","title":"Optimal Control Theoretic Neural Optimizer: From Backpropagation to Dynamic Programming","display_name":"Optimal Control Theoretic Neural Optimizer: From Backpropagation to Dynamic Programming","publication_year":2025,"publication_date":"2025-11-19","ids":{"openalex":"https://openalex.org/W4416366417","doi":"https://doi.org/10.1109/tpami.2025.3634391","pmid":"https://pubmed.ncbi.nlm.nih.gov/41259159"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2025.3634391","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2025.3634391","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031381142","display_name":"Guan-Horng Liu","orcid":"https://orcid.org/0000-0002-8989-7568"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Guan-Horng Liu","raw_affiliation_strings":["FAIR, Meta, Menlo Park, CA, USA","FAIR, Meta"],"raw_orcid":"https://orcid.org/0000-0002-8989-7568","affiliations":[{"raw_affiliation_string":"FAIR, Meta, Menlo Park, CA, USA","institution_ids":[]},{"raw_affiliation_string":"FAIR, Meta","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087793625","display_name":"Tianrong Chen","orcid":"https://orcid.org/0000-0003-1701-2991"},"institutions":[{"id":"https://openalex.org/I1311269955","display_name":"Apple (Israel)","ror":"https://ror.org/04ehjr030","country_code":"IL","type":"company","lineage":["https://openalex.org/I1311269955","https://openalex.org/I4210153776"]},{"id":"https://openalex.org/I4210153776","display_name":"Apple (United States)","ror":"https://ror.org/059hsda18","country_code":"US","type":"company","lineage":["https://openalex.org/I4210153776"]}],"countries":["IL","US"],"is_corresponding":false,"raw_author_name":"Tianrong Chen","raw_affiliation_strings":["Apple MLR, Cupertino, CA, USA","Apple MLR"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Apple MLR, Cupertino, CA, USA","institution_ids":["https://openalex.org/I4210153776"]},{"raw_affiliation_string":"Apple MLR","institution_ids":["https://openalex.org/I1311269955","https://openalex.org/I4210153776"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102986128","display_name":"Evangelos Theodorou","orcid":"https://orcid.org/0000-0002-8270-2473"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]},{"id":"https://openalex.org/I4210161101","display_name":"Research Institute of Technology (Russia)","ror":"https://ror.org/056acjt11","country_code":"RU","type":"company","lineage":["https://openalex.org/I4210161101"]}],"countries":["RU","US"],"is_corresponding":false,"raw_author_name":"Evangelos A. Theodorou","raw_affiliation_strings":["School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, Georgia","Aerospace EngineeringGeorgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, Georgia","institution_ids":["https://openalex.org/I130701444"]},{"raw_affiliation_string":"Aerospace EngineeringGeorgia Institute of Technology","institution_ids":["https://openalex.org/I4210161101"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5031381142"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.18117463,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"48","issue":"3","first_page":"2989","last_page":"3001"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.20180000364780426,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.20180000364780426,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11206","display_name":"Model Reduction and Neural Networks","score":0.1914999932050705,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.14740000665187836,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.7896000146865845},{"id":"https://openalex.org/keywords/optimal-control","display_name":"Optimal control","score":0.7074000239372253},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6437000036239624},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.635699987411499},{"id":"https://openalex.org/keywords/dynamic-programming","display_name":"Dynamic programming","score":0.603600025177002},{"id":"https://openalex.org/keywords/dynamical-systems-theory","display_name":"Dynamical systems theory","score":0.5658000111579895},{"id":"https://openalex.org/keywords/nonlinear-system","display_name":"Nonlinear system","score":0.42010000348091125},{"id":"https://openalex.org/keywords/optimization-problem","display_name":"Optimization problem","score":0.3849000036716461}],"concepts":[{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.7896000146865845},{"id":"https://openalex.org/C91575142","wikidata":"https://www.wikidata.org/wiki/Q1971426","display_name":"Optimal control","level":2,"score":0.7074000239372253},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6460999846458435},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6437000036239624},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.635699987411499},{"id":"https://openalex.org/C37404715","wikidata":"https://www.wikidata.org/wiki/Q380679","display_name":"Dynamic programming","level":2,"score":0.603600025177002},{"id":"https://openalex.org/C79379906","wikidata":"https://www.wikidata.org/wiki/Q3174497","display_name":"Dynamical systems theory","level":2,"score":0.5658000111579895},{"id":"https://openalex.org/C158622935","wikidata":"https://www.wikidata.org/wiki/Q660848","display_name":"Nonlinear system","level":2,"score":0.42010000348091125},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4171999990940094},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.39649999141693115},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.3849000036716461},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.3650999963283539},{"id":"https://openalex.org/C33962884","wikidata":"https://www.wikidata.org/wiki/Q378637","display_name":"Dynamical system (definition)","level":3,"score":0.3483999967575073},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3276999890804291},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32359999418258667},{"id":"https://openalex.org/C115527620","wikidata":"https://www.wikidata.org/wiki/Q769909","display_name":"Nonlinear programming","level":3,"score":0.3098999857902527},{"id":"https://openalex.org/C31531917","wikidata":"https://www.wikidata.org/wiki/Q915157","display_name":"Robust control","level":3,"score":0.30880001187324524},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2953999936580658},{"id":"https://openalex.org/C177973122","wikidata":"https://www.wikidata.org/wiki/Q7860946","display_name":"Types of artificial neural networks","level":4,"score":0.2757999897003174},{"id":"https://openalex.org/C17500928","wikidata":"https://www.wikidata.org/wiki/Q959968","display_name":"Control system","level":2,"score":0.2743000090122223},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C114275822","wikidata":"https://www.wikidata.org/wiki/Q621512","display_name":"Linear dynamical system","level":3,"score":0.26989999413490295},{"id":"https://openalex.org/C77405623","wikidata":"https://www.wikidata.org/wiki/Q598451","display_name":"System dynamics","level":2,"score":0.257999986410141},{"id":"https://openalex.org/C98359873","wikidata":"https://www.wikidata.org/wiki/Q1320470","display_name":"Rprop","level":5,"score":0.2556999921798706}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2025.3634391","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2025.3634391","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:41259159","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/41259159","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320309321","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W38953637","https://openalex.org/W1587799944","https://openalex.org/W1978099898","https://openalex.org/W1980516134","https://openalex.org/W2055381865","https://openalex.org/W2056261501","https://openalex.org/W2087617385","https://openalex.org/W2144144709","https://openalex.org/W2167856595","https://openalex.org/W2194775991","https://openalex.org/W2296226496","https://openalex.org/W2558748708","https://openalex.org/W2600297185","https://openalex.org/W2811004130","https://openalex.org/W2898917953","https://openalex.org/W2919115771","https://openalex.org/W2963150697","https://openalex.org/W2963240808","https://openalex.org/W2963359731","https://openalex.org/W3004424267","https://openalex.org/W3133294546","https://openalex.org/W4302010773"],"related_works":[],"abstract_inverted_index":{"Optimization":[0],"of":[1,21,40,127,133,159],"deep":[2],"neural":[3],"networks":[4],"(DNNs)":[5],"has":[6,45],"been":[7],"driving":[8],"modern":[9],"advancements":[10],"in":[11,48,85,175,192],"artificial":[12],"intelligence.":[13],"With":[14],"DNNs":[15,41,86],"characterized":[16],"by":[17],"a":[18,109,124],"prolonged":[19],"sequence":[20],"nonlinear":[22],"propagation,":[23],"determining":[24],"their":[25],"optimal":[26,196],"parameters":[27],"given":[28],"an":[29,38,67,113],"objective":[30],"naturally":[31],"fits":[32],"within":[33],"Optimal":[34,140],"Control":[35,141],"Programming.":[36],"Such":[37],"interpretation":[39],"as":[42,100,163],"dynamical":[43,92,193],"systems":[44,194],"proven":[46],"crucial":[47],"offering":[49],"principled":[50,188],"analysis":[51],"from":[52],"numerical":[53],"equations":[54],"to":[55,59,118,123],"physics.":[56],"In":[57],"parallel":[58],"these":[60],"theoretical":[61],"pursuits,":[62],"this":[63,104],"paper":[64],"focuses":[65],"on":[66],"algorithmic":[68,76,148,189],"perspective.":[69],"Our":[70],"motivated":[71],"observation":[72],"is":[73],"the":[74,79,88,119,134],"striking":[75],"resemblance":[77],"between":[78],"Backpropagation":[80,107],"algorithm":[81],"for":[82,91,187],"computing":[83],"gradients":[84],"and":[87,156,177,195],"optimality":[89],"conditions":[90],"systems,":[93],"expressed":[94],"through":[95],"another":[96],"backward":[97],"process":[98],"known":[99],"dynamic":[101,115],"programming.":[102],"Consolidating":[103],"connection,":[105],"where":[106],"admits":[108],"variational":[110],"structure,":[111],"solving":[112],"approximate":[114],"programming":[116],"up":[117],"first-order":[120],"expansion,":[121],"leads":[122],"new":[125,185],"class":[126],"optimization":[128],"methods":[129,174],"exploring":[130],"higher-order":[131,157],"expansions":[132],"Bellman":[135],"equation.":[136],"The":[137],"resulting":[138],"optimizer,":[139],"Theoretic":[142],"Neural":[143,164],"Optimizer":[144],"(OCNOpt),":[145],"enables":[146],"rich":[147],"opportunities,":[149],"including":[150],"layer-wise":[151],"feedback":[152],"policies,":[153],"game-theoretic":[154],"applications,":[155],"training":[158],"continuous-time":[160],"models":[161],"such":[162],"ODEs.":[165],"Extensive":[166],"experiments":[167],"demonstrate":[168],"that":[169],"OCNOpt":[170],"improves":[171],"upon":[172],"existing":[173],"robustness":[176],"efficiency":[178],"while":[179],"maintaining":[180],"manageable":[181],"computational":[182],"complexity,":[183],"paving":[184],"avenues":[186],"design":[190],"grounded":[191],"control":[197],"theory.":[198]},"counts_by_year":[],"updated_date":"2026-02-07T06:11:34.122080","created_date":"2025-11-19T00:00:00"}