{"id":"https://openalex.org/W3159197609","doi":"https://doi.org/10.1109/icpr48806.2021.9412188","title":"RNN Training along Locally Optimal Trajectories via Frank-Wolfe Algorithm","display_name":"RNN Training along Locally Optimal Trajectories via Frank-Wolfe Algorithm","publication_year":2021,"publication_date":"2021-01-10","ids":{"openalex":"https://openalex.org/W3159197609","doi":"https://doi.org/10.1109/icpr48806.2021.9412188","mag":"3159197609"},"language":"en","primary_location":{"id":"doi:10.1109/icpr48806.2021.9412188","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpr48806.2021.9412188","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 25th International Conference on Pattern Recognition (ICPR)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"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/A5112262878","display_name":"Yun Yue","orcid":null},"institutions":[{"id":"https://openalex.org/I107077323","display_name":"Worcester Polytechnic Institute","ror":"https://ror.org/05ejpqr48","country_code":"US","type":"education","lineage":["https://openalex.org/I107077323"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yun Yue","raw_affiliation_strings":["Worcester Polytechnic Institute, Worcester, MA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Worcester Polytechnic Institute, Worcester, MA","institution_ids":["https://openalex.org/I107077323"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100351468","display_name":"Ming Li","orcid":"https://orcid.org/0000-0002-7852-0159"},"institutions":[{"id":"https://openalex.org/I107077323","display_name":"Worcester Polytechnic Institute","ror":"https://ror.org/05ejpqr48","country_code":"US","type":"education","lineage":["https://openalex.org/I107077323"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ming Li","raw_affiliation_strings":["Worcester Polytechnic Institute, Worcester, MA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Worcester Polytechnic Institute, Worcester, MA","institution_ids":["https://openalex.org/I107077323"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048704387","display_name":"Venkatesh Saligrama","orcid":"https://orcid.org/0000-0002-0675-2268"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Venkatesh Saligrama","raw_affiliation_strings":["Boston University, Boston, MA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Boston University, Boston, MA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5048506332","display_name":"Ziming Zhang","orcid":"https://orcid.org/0000-0002-0921-8298"},"institutions":[{"id":"https://openalex.org/I107077323","display_name":"Worcester Polytechnic Institute","ror":"https://ror.org/05ejpqr48","country_code":"US","type":"education","lineage":["https://openalex.org/I107077323"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ziming Zhang","raw_affiliation_strings":["Worcester Polytechnic Institute, Worcester, MA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Worcester Polytechnic Institute, Worcester, MA","institution_ids":["https://openalex.org/I107077323"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":"70","issue":null,"first_page":"10532","last_page":"10539"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9997000098228455,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9997000098228455,"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/T12676","display_name":"Machine Learning and ELM","score":0.9993000030517578,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9961000084877014,"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/leverage","display_name":"Leverage (statistics)","score":0.7372952699661255},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6831914186477661},{"id":"https://openalex.org/keywords/maxima-and-minima","display_name":"Maxima and minima","score":0.6582980155944824},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.6308488249778748},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.5555852055549622},{"id":"https://openalex.org/keywords/sublinear-function","display_name":"Sublinear function","score":0.543419599533081},{"id":"https://openalex.org/keywords/rate-of-convergence","display_name":"Rate of convergence","score":0.5254713296890259},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.49238288402557373},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4694475829601288},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.4604552090167999},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.45290476083755493},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.37484705448150635},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.339174747467041},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.32513269782066345},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.23054009675979614},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.13712400197982788}],"concepts":[{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.7372952699661255},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6831914186477661},{"id":"https://openalex.org/C186633575","wikidata":"https://www.wikidata.org/wiki/Q845060","display_name":"Maxima and minima","level":2,"score":0.6582980155944824},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6308488249778748},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5555852055549622},{"id":"https://openalex.org/C117160843","wikidata":"https://www.wikidata.org/wiki/Q338652","display_name":"Sublinear function","level":2,"score":0.543419599533081},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.5254713296890259},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.49238288402557373},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4694475829601288},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.4604552090167999},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.45290476083755493},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.37484705448150635},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.339174747467041},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.32513269782066345},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.23054009675979614},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.13712400197982788},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icpr48806.2021.9412188","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpr48806.2021.9412188","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 25th International Conference on Pattern Recognition (ICPR)","raw_type":"proceedings-article"},{"id":"pmh:oai:null:2144/44092","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ICPR48806.2021.9412188","pdf_url":null,"source":{"id":"https://openalex.org/S4306402384","display_name":"OpenBU (Boston University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I111088046","host_organization_name":"Boston University","host_organization_lineage":["https://openalex.org/I111088046"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"Conference materials"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6936287966","display_name":null,"funder_award_id":"CCF-2006738","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":113,"referenced_works":["https://openalex.org/W905619","https://openalex.org/W1482804306","https://openalex.org/W1522301498","https://openalex.org/W1574851760","https://openalex.org/W1771410628","https://openalex.org/W1800356822","https://openalex.org/W1815076433","https://openalex.org/W1981039871","https://openalex.org/W2001263627","https://openalex.org/W2008308152","https://openalex.org/W2016589492","https://openalex.org/W2038904236","https://openalex.org/W2040055413","https://openalex.org/W2062210787","https://openalex.org/W2064675550","https://openalex.org/W2087831875","https://openalex.org/W2108563286","https://openalex.org/W2112246162","https://openalex.org/W2128084896","https://openalex.org/W2135473110","https://openalex.org/W2136885855","https://openalex.org/W2140575957","https://openalex.org/W2175402905","https://openalex.org/W2243249356","https://openalex.org/W2547718140","https://openalex.org/W2557270725","https://openalex.org/W2557389081","https://openalex.org/W2581719241","https://openalex.org/W2587753047","https://openalex.org/W2618854269","https://openalex.org/W2619808423","https://openalex.org/W2620263509","https://openalex.org/W2750814024","https://openalex.org/W2752813595","https://openalex.org/W2753358588","https://openalex.org/W2790155499","https://openalex.org/W2890309926","https://openalex.org/W2898324627","https://openalex.org/W2906125418","https://openalex.org/W2908046483","https://openalex.org/W2914623577","https://openalex.org/W2918846640","https://openalex.org/W2942751004","https://openalex.org/W2950059857","https://openalex.org/W2950297649","https://openalex.org/W2951605425","https://openalex.org/W2952436057","https://openalex.org/W2958796454","https://openalex.org/W2962840019","https://openalex.org/W2962867868","https://openalex.org/W2963010571","https://openalex.org/W2963042606","https://openalex.org/W2963052924","https://openalex.org/W2963167419","https://openalex.org/W2963174729","https://openalex.org/W2963241221","https://openalex.org/W2963276946","https://openalex.org/W2963755523","https://openalex.org/W2963983719","https://openalex.org/W2963992014","https://openalex.org/W2964121744","https://openalex.org/W2964182247","https://openalex.org/W2964199361","https://openalex.org/W2964269252","https://openalex.org/W2964304579","https://openalex.org/W2964335273","https://openalex.org/W2964347220","https://openalex.org/W2970418991","https://openalex.org/W2991290085","https://openalex.org/W2994747431","https://openalex.org/W2999543292","https://openalex.org/W3005560345","https://openalex.org/W4250589301","https://openalex.org/W4288257325","https://openalex.org/W4294555862","https://openalex.org/W4294568934","https://openalex.org/W6600043445","https://openalex.org/W6631190155","https://openalex.org/W6638018090","https://openalex.org/W6638114406","https://openalex.org/W6638545294","https://openalex.org/W6639497327","https://openalex.org/W6676018521","https://openalex.org/W6685537299","https://openalex.org/W6690321021","https://openalex.org/W6691800262","https://openalex.org/W6720905350","https://openalex.org/W6729305679","https://openalex.org/W6729992640","https://openalex.org/W6730235532","https://openalex.org/W6732435028","https://openalex.org/W6733068669","https://openalex.org/W6738195075","https://openalex.org/W6738420126","https://openalex.org/W6743485176","https://openalex.org/W6743554579","https://openalex.org/W6743969146","https://openalex.org/W6744262642","https://openalex.org/W6748380982","https://openalex.org/W6748961382","https://openalex.org/W6752307458","https://openalex.org/W6754160923","https://openalex.org/W6757170499","https://openalex.org/W6757928074","https://openalex.org/W6758805728","https://openalex.org/W6760181056","https://openalex.org/W6762509139","https://openalex.org/W6764398373","https://openalex.org/W6765671628","https://openalex.org/W6766786322","https://openalex.org/W6772129282","https://openalex.org/W6772509012","https://openalex.org/W6844220378"],"related_works":["https://openalex.org/W1535694475","https://openalex.org/W1558978786","https://openalex.org/W1965562977","https://openalex.org/W2135804779","https://openalex.org/W2093953062","https://openalex.org/W1554143855","https://openalex.org/W2945629716","https://openalex.org/W4283775266","https://openalex.org/W3043533097","https://openalex.org/W4298860769"],"abstract_inverted_index":{"We":[0,35,105,126,146],"propose":[1,36],"a":[2,13,21,56,62,90,102,114],"novel":[3,63],"and":[4,24,141,153],"efficient":[5,155],"training":[6,65,76,156,163],"method":[7,66,87,164],"for":[8,29,123],"RNNs":[9],"by":[10],"iteratively":[11],"seeking":[12],"local":[14],"minima":[15],"on":[16,131],"the":[17,30,39,71,74],"loss":[18],"surface":[19],"within":[20],"small":[22],"region,":[23],"leverage":[25],"this":[26,44],"directional":[27],"vector":[28],"update,":[31],"in":[32,43,96],"an":[33,98],"outer-loop.":[34],"to":[37,55,81,89,167],"utilize":[38],"Frank-Wolfe":[40,92],"(FW)":[41],"algorithm":[42,100,112],"context.":[45],"Although,":[46],"FW":[47],"implicitly":[48],"involves":[49],"normalized":[50],"gradients,":[51],"which":[52],"can":[53],"lead":[54],"slow":[57],"convergence":[58,116],"rate,":[59],"we":[60,159],"develop":[61],"RNN":[64,151],"that,":[67],"surprisingly,":[68],"even":[69],"with":[70,101,149],"additional":[72],"cost,":[73],"overall":[75],"cost":[77],"is":[78,95,165],"empirically":[79],"observed":[80],"be":[82],"lower":[83],"than":[84],"backpropagation.":[85],"Our":[86],"leads":[88],"new":[91],"method,":[93],"that":[94,107,137,161],"essence":[97],"SGD":[99],"restart":[103],"scheme.":[104],"prove":[106],"under":[108],"certain":[109],"conditions":[110],"our":[111,162],"has":[113],"sublinear":[115],"rate":[117],"of":[118],"<i":[119],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[120],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">O</i>":[121],"(1/\u03f5)":[122],"\u03f5":[124],"error.":[125],"then":[127],"conduct":[128],"empirical":[129],"experiments":[130],"several":[132],"benchmark":[133],"datasets":[134],"including":[135],"those":[136],"exhibit":[138],"long-term":[139],"dependencies,":[140],"show":[142,154],"significant":[143],"performance":[144],"improvement.":[145],"also":[147],"experiment":[148],"deep":[150],"architectures":[152],"performance.":[157],"Finally,":[158],"demonstrate":[160],"robust":[166],"noisy":[168],"data.":[169]},"counts_by_year":[{"year":2023,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}