{"id":"https://openalex.org/W4415624367","doi":"https://doi.org/10.1109/tnnls.2025.3620798","title":"Progressive Feedforward Collapse of ResNet Training","display_name":"Progressive Feedforward Collapse of ResNet Training","publication_year":2025,"publication_date":"2025-10-28","ids":{"openalex":"https://openalex.org/W4415624367","doi":"https://doi.org/10.1109/tnnls.2025.3620798","pmid":"https://pubmed.ncbi.nlm.nih.gov/41150251"},"language":"en","primary_location":{"id":"doi:10.1109/tnnls.2025.3620798","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnnls.2025.3620798","pdf_url":null,"source":{"id":"https://openalex.org/S4210175523","display_name":"IEEE Transactions on Neural Networks and Learning Systems","issn_l":"2162-237X","issn":["2162-237X","2162-2388"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["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 Neural Networks and Learning Systems","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/A5109866180","display_name":"Sicong Wang","orcid":"https://orcid.org/0009-0000-0757-278X"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210120485","display_name":"Academy of Mathematics and Systems Science","ror":"https://ror.org/02jkmyk67","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210120485"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Sicong Wang","raw_affiliation_strings":["Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0000-0757-278X","affiliations":[{"raw_affiliation_string":"Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210120485","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103110172","display_name":"Kuo Gai","orcid":"https://orcid.org/0009-0004-7169-4414"},"institutions":[{"id":"https://openalex.org/I4403928334","display_name":"Shanghai Institute for Mathematics and Interdisciplinary Sciences","ror":"https://ror.org/0489cd473","country_code":null,"type":"government","lineage":["https://openalex.org/I4403928334"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kuo Gai","raw_affiliation_strings":["Shanghai Institute for Mathematics and Interdisciplinary Sciences, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0004-7169-4414","affiliations":[{"raw_affiliation_string":"Shanghai Institute for Mathematics and Interdisciplinary Sciences, Shanghai, China","institution_ids":["https://openalex.org/I4403928334"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5076619069","display_name":"Shihua Zhang","orcid":"https://orcid.org/0000-0003-0192-7118"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210120485","display_name":"Academy of Mathematics and Systems Science","ror":"https://ror.org/02jkmyk67","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210120485"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shihua Zhang","raw_affiliation_strings":["Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-0192-7118","affiliations":[{"raw_affiliation_string":"Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210120485","https://openalex.org/I19820366"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.15626438,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"37","issue":"3","first_page":"1078","last_page":"1091"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.5223000049591064,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.5223000049591064,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.09139999747276306,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.06750000268220901,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5726000070571899},{"id":"https://openalex.org/keywords/geodesic","display_name":"Geodesic","score":0.49140000343322754},{"id":"https://openalex.org/keywords/control-theory","display_name":"Control theory (sociology)","score":0.439300000667572},{"id":"https://openalex.org/keywords/residual-neural-network","display_name":"Residual neural network","score":0.39959999918937683},{"id":"https://openalex.org/keywords/monotonic-function","display_name":"Monotonic function","score":0.39089998602867126},{"id":"https://openalex.org/keywords/simplex","display_name":"Simplex","score":0.3831999897956848},{"id":"https://openalex.org/keywords/feedforward-neural-network","display_name":"Feedforward neural network","score":0.37950000166893005},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.35589998960494995}],"concepts":[{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5726000070571899},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5094000101089478},{"id":"https://openalex.org/C165818556","wikidata":"https://www.wikidata.org/wiki/Q213488","display_name":"Geodesic","level":2,"score":0.49140000343322754},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4634000062942505},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.439300000667572},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.41909998655319214},{"id":"https://openalex.org/C2944601119","wikidata":"https://www.wikidata.org/wiki/Q43744058","display_name":"Residual neural network","level":3,"score":0.39959999918937683},{"id":"https://openalex.org/C72169020","wikidata":"https://www.wikidata.org/wiki/Q194404","display_name":"Monotonic function","level":2,"score":0.39089998602867126},{"id":"https://openalex.org/C62438384","wikidata":"https://www.wikidata.org/wiki/Q331350","display_name":"Simplex","level":2,"score":0.3831999897956848},{"id":"https://openalex.org/C47702885","wikidata":"https://www.wikidata.org/wiki/Q5441227","display_name":"Feedforward neural network","level":3,"score":0.37950000166893005},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3560999929904938},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.35589998960494995},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.32499998807907104},{"id":"https://openalex.org/C38858127","wikidata":"https://www.wikidata.org/wiki/Q5441228","display_name":"Feed forward","level":2,"score":0.32429999113082886},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.3190000057220459},{"id":"https://openalex.org/C144521790","wikidata":"https://www.wikidata.org/wiki/Q134164","display_name":"Simplex algorithm","level":3,"score":0.3176000118255615},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.2800999879837036},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.27399998903274536},{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.2736000120639801},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.2621000111103058}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tnnls.2025.3620798","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnnls.2025.3620798","pdf_url":null,"source":{"id":"https://openalex.org/S4210175523","display_name":"IEEE Transactions on Neural Networks and Learning Systems","issn_l":"2162-237X","issn":["2162-237X","2162-2388"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["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 Neural Networks and Learning Systems","raw_type":"journal-article"},{"id":"pmid:41150251","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/41150251","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 neural networks and learning systems","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2298682283","display_name":null,"funder_award_id":"12326614","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8309094176","display_name":null,"funder_award_id":"32341013","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1995562189","https://openalex.org/W2194775991","https://openalex.org/W2302255633","https://openalex.org/W2566079294","https://openalex.org/W2612252966","https://openalex.org/W2963518130","https://openalex.org/W2964184826","https://openalex.org/W2989653627","https://openalex.org/W3004591935","https://openalex.org/W3042925017","https://openalex.org/W3065974826","https://openalex.org/W3109311007","https://openalex.org/W3134837785","https://openalex.org/W3205278400","https://openalex.org/W4206487645","https://openalex.org/W4226192560","https://openalex.org/W4252085168","https://openalex.org/W4315641811","https://openalex.org/W4385864301","https://openalex.org/W4386211960","https://openalex.org/W4395661445","https://openalex.org/W4401507728"],"related_works":[],"abstract_inverted_index":{"Neural":[0],"collapse":[1,24,81,87,184],"(NC)":[2],"is":[3,161,166],"a":[4,31,76,97,139],"simple":[5],"and":[6,29,52,74,189],"symmetric":[7],"phenomenon":[8,185],"for":[9,100],"deep":[10],"neural":[11,71],"networks":[12],"(DNNs)":[13],"at":[14,121],"the":[15,21,39,43,46,50,64,84,90,101,106,114,118,122,171,183,193,199],"terminal":[16,123],"phase":[17],"of":[18,45,66,69,86,93,127,159,186,202],"training,":[19],"where":[20],"last-layer":[22,47],"features":[23,48],"to":[25,49,170,179,181],"their":[26],"class":[27],"means":[28],"form":[30],"simplex":[32],"equiangular":[33],"tight":[34],"frame":[35],"(ETF)":[36],"aligning":[37],"with":[38,110,163],"classifier":[40],"vectors.":[41],"However,":[42],"relationship":[44],"data":[51],"intermediate":[53,67,149,187],"layers":[54,68,150,188],"during":[55,89],"training":[56],"remains":[57],"unexplored.":[58],"To":[59],"this":[60,175],"end,":[61],"we":[62],"characterize":[63],"geometry":[65],"residual":[70],"network":[72],"(ResNet)":[73],"propose":[75,138],"novel":[77],"conjecture,":[78],"progressive":[79],"feedforward":[80],"(PFC),":[82],"claiming":[83],"degree":[85],"increases":[88],"forward":[91],"propagation":[92],"DNNs.":[94],"We":[95,137],"derive":[96],"transparent":[98],"model":[99,146,182],"well-trained":[102],"ResNet":[103,109,203],"based":[104],"on":[105,134,192,198],"principle":[107],"that":[108],"weight":[111],"decay":[112],"approximates":[113],"geodesic":[115],"curve":[116],"in":[117,204],"Wasserstein":[119],"space":[120],"phase.":[124],"The":[125,156],"metrics":[126],"PFC":[128,180],"indeed":[129],"monotonically":[130],"decrease":[131],"across":[132],"depth":[133],"various":[135],"datasets.":[136],"new":[140],"surrogate":[141],"model,":[142],"multilayer":[143],"unconstrained":[144],"feature":[145],"(MUFM),":[147],"connecting":[148],"by":[151],"an":[152],"optimal":[153,157],"transport":[154],"regularizer.":[155],"solution":[158],"MUFM":[160],"inconsistent":[162],"NC":[164,178],"but":[165],"more":[167],"concentrated":[168],"relative":[169],"input":[172,194],"data.":[173],"Overall,":[174],"study":[176],"extends":[177],"its":[190],"dependence":[191],"data,":[195],"shedding":[196],"light":[197],"theoretical":[200],"understanding":[201],"classification":[205],"problems.":[206]},"counts_by_year":[],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-28T00:00:00"}