{"id":"https://openalex.org/W4402389513","doi":"https://doi.org/10.1007/s10957-024-02513-3","title":"Gradient Descent Provably Escapes Saddle Points in the Training of Shallow ReLU Networks","display_name":"Gradient Descent Provably Escapes Saddle Points in the Training of Shallow ReLU Networks","publication_year":2024,"publication_date":"2024-09-10","ids":{"openalex":"https://openalex.org/W4402389513","doi":"https://doi.org/10.1007/s10957-024-02513-3","pmid":"https://pubmed.ncbi.nlm.nih.gov/39664032"},"language":"en","primary_location":{"id":"doi:10.1007/s10957-024-02513-3","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10957-024-02513-3","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10957-024-02513-3.pdf","source":{"id":"https://openalex.org/S52430896","display_name":"Journal of Optimization Theory and Applications","issn_l":"0022-3239","issn":["0022-3239","1573-2878"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Optimization Theory and Applications","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s10957-024-02513-3.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5002005113","display_name":"Patrick Cheridito","orcid":"https://orcid.org/0000-0001-9074-7295"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":true,"raw_author_name":"Patrick Cheridito","raw_affiliation_strings":["Department of Mathematics and RiskLab, ETH Zurich, Zurich, Switzerland","Department of Mathematics and RiskLab, ETH Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"Department of Mathematics and RiskLab, ETH Zurich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]},{"raw_affiliation_string":"Department of Mathematics and RiskLab, ETH Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074382323","display_name":"Arnulf Jentzen","orcid":"https://orcid.org/0000-0002-9840-3339"},"institutions":[{"id":"https://openalex.org/I22465464","display_name":"University of M\u00fcnster","ror":"https://ror.org/00pd74e08","country_code":"DE","type":"education","lineage":["https://openalex.org/I22465464"]},{"id":"https://openalex.org/I4210099586","display_name":"Shenzhen Research Institute of Big Data","ror":"https://ror.org/00z1gwf89","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210099586"]},{"id":"https://openalex.org/I4210116924","display_name":"Chinese University of Hong Kong, Shenzhen","ror":"https://ror.org/02d5ks197","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633","https://openalex.org/I180726961","https://openalex.org/I4210116924"]}],"countries":["CN","DE"],"is_corresponding":false,"raw_author_name":"Arnulf Jentzen","raw_affiliation_strings":["Applied Mathematics: Institute for Analysis and Numerics, University of M\u00fcnster, M\u00fcnster, Germany","School of Data Science and Shenzhen Research Institute of Big Data, The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), Shenzhen, China","School of Data Science and Shenzhen Research Institute of Big Data, The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), China","Applied Mathematics: Institute for Analysis and Numerics, University of M\u00fcnster, Germany"],"affiliations":[{"raw_affiliation_string":"Applied Mathematics: Institute for Analysis and Numerics, University of M\u00fcnster, M\u00fcnster, Germany","institution_ids":["https://openalex.org/I22465464"]},{"raw_affiliation_string":"School of Data Science and Shenzhen Research Institute of Big Data, The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), Shenzhen, China","institution_ids":["https://openalex.org/I4210116924","https://openalex.org/I4210099586"]},{"raw_affiliation_string":"School of Data Science and Shenzhen Research Institute of Big Data, The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), China","institution_ids":["https://openalex.org/I4210116924","https://openalex.org/I4210099586"]},{"raw_affiliation_string":"Applied Mathematics: Institute for Analysis and Numerics, University of M\u00fcnster, Germany","institution_ids":["https://openalex.org/I22465464"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001762952","display_name":"Florian Rossmannek","orcid":"https://orcid.org/0000-0001-5772-5086"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]},{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH","SG"],"is_corresponding":false,"raw_author_name":"Florian Rossmannek","raw_affiliation_strings":["Department of Mathematics and RiskLab, ETH Zurich, Zurich, Switzerland","School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore, Singapore","Department of Mathematics and RiskLab, ETH Zurich, Switzerland","School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore"],"affiliations":[{"raw_affiliation_string":"Department of Mathematics and RiskLab, ETH Zurich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]},{"raw_affiliation_string":"School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore, Singapore","institution_ids":["https://openalex.org/I172675005"]},{"raw_affiliation_string":"Department of Mathematics and RiskLab, ETH Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]},{"raw_affiliation_string":"School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5002005113"],"corresponding_institution_ids":["https://openalex.org/I35440088"],"apc_list":{"value":2390,"currency":"EUR","value_usd":2990},"apc_paid":{"value":2390,"currency":"EUR","value_usd":2990},"fwci":0.3491,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.66252913,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":"203","issue":"3","first_page":"2617","last_page":"2648"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9980000257492065,"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":0.9980000257492065,"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/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9979000091552734,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12676","display_name":"Machine Learning and ELM","score":0.994700014591217,"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/saddle-point","display_name":"Saddle point","score":0.838009774684906},{"id":"https://openalex.org/keywords/theory-of-computation","display_name":"Theory of computation","score":0.7349393963813782},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.7135043144226074},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.6557841300964355},{"id":"https://openalex.org/keywords/saddle","display_name":"Saddle","score":0.6146817803382874},{"id":"https://openalex.org/keywords/descent","display_name":"Descent (aeronautics)","score":0.6083243489265442},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5421233773231506},{"id":"https://openalex.org/keywords/descent-direction","display_name":"Descent direction","score":0.4396052062511444},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.4210321605205536},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.3637451231479645},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.32170259952545166},{"id":"https://openalex.org/keywords/geometry","display_name":"Geometry","score":0.2868383526802063},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.24334785342216492},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.23739734292030334},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.2320224940776825},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.12460371851921082},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.08728975057601929}],"concepts":[{"id":"https://openalex.org/C2681867","wikidata":"https://www.wikidata.org/wiki/Q690935","display_name":"Saddle point","level":2,"score":0.838009774684906},{"id":"https://openalex.org/C24858836","wikidata":"https://www.wikidata.org/wiki/Q844718","display_name":"Theory of computation","level":2,"score":0.7349393963813782},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.7135043144226074},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.6557841300964355},{"id":"https://openalex.org/C2777127463","wikidata":"https://www.wikidata.org/wiki/Q10862618","display_name":"Saddle","level":2,"score":0.6146817803382874},{"id":"https://openalex.org/C2776637919","wikidata":"https://www.wikidata.org/wiki/Q624380","display_name":"Descent (aeronautics)","level":2,"score":0.6083243489265442},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5421233773231506},{"id":"https://openalex.org/C116149140","wikidata":"https://www.wikidata.org/wiki/Q2070951","display_name":"Descent direction","level":4,"score":0.4396052062511444},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.4210321605205536},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3637451231479645},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.32170259952545166},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.2868383526802063},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.24334785342216492},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.23739734292030334},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.2320224940776825},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.12460371851921082},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.08728975057601929},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1007/s10957-024-02513-3","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10957-024-02513-3","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10957-024-02513-3.pdf","source":{"id":"https://openalex.org/S52430896","display_name":"Journal of Optimization Theory and Applications","issn_l":"0022-3239","issn":["0022-3239","1573-2878"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Optimization Theory and Applications","raw_type":"journal-article"},{"id":"pmid:39664032","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/39664032","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":"Journal of optimization theory and applications","raw_type":null},{"id":"pmh:oai:pubmedcentral.nih.gov:11628594","is_oa":true,"landing_page_url":"https://www.ncbi.nlm.nih.gov/pmc/articles/11628594","pdf_url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC11628594/pdf/10957_2024_Article_2513.pdf","source":{"id":"https://openalex.org/S2764455111","display_name":"PubMed Central","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"J Optim Theory Appl","raw_type":"Text"},{"id":"pmh:oai:RePEc:spr:joptap:v:203:y:2024:i:3:d:10.1007_s10957-024-02513-3","is_oa":false,"landing_page_url":"http://link.springer.com/10.1007/s10957-024-02513-3","pdf_url":null,"source":{"id":"https://openalex.org/S4306401271","display_name":"RePEc: Research Papers in Economics","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I77793887","host_organization_name":"Federal Reserve Bank of St. Louis","host_organization_lineage":["https://openalex.org/I77793887"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.1007/s10957-024-02513-3","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10957-024-02513-3","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10957-024-02513-3.pdf","source":{"id":"https://openalex.org/S52430896","display_name":"Journal of Optimization Theory and Applications","issn_l":"0022-3239","issn":["0022-3239","1573-2878"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Optimization Theory and Applications","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1177974802","display_name":null,"funder_award_id":"EXC 2044-390685587","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G1652504919","display_name":null,"funder_award_id":"EXC 2044","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G2110285270","display_name":null,"funder_award_id":"2044-390685587","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G2817606592","display_name":null,"funder_award_id":"390685587","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G4263190246","display_name":null,"funder_award_id":"101045811","funder_id":"https://openalex.org/F4320338453","funder_display_name":"HORIZON EUROPE European Research Council"},{"id":"https://openalex.org/G5106512922","display_name":null,"funder_award_id":"Deutsche Forschungsgemeinschaft (DFG","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G670523356","display_name":null,"funder_award_id":"Germany\u2019s Excellence Strategy EXC 2044-390685587","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G6743243744","display_name":null,"funder_award_id":"unknown","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G8330397625","display_name":null,"funder_award_id":"Funding","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G8526424774","display_name":null,"funder_award_id":"Fellowship","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G870869636","display_name":null,"funder_award_id":"EXC 2044-39068558","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G8724757341","display_name":"Overcoming the curse of dimensionality through nonlinear stochastic algorithms","funder_award_id":"101045811","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G8928656450","display_name":null,"funder_award_id":"Projekt DEAL","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"},{"id":"https://openalex.org/F4320322957","display_name":"Westf\u00e4lische Wilhelms-Universit\u00e4t M\u00fcnster","ror":"https://ror.org/00pd74e08"},{"id":"https://openalex.org/F4320338453","display_name":"HORIZON EUROPE European Research Council","ror":null}],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4402389513.pdf"},"referenced_works_count":23,"referenced_works":["https://openalex.org/W1550918041","https://openalex.org/W2004160833","https://openalex.org/W2015836240","https://openalex.org/W2077612096","https://openalex.org/W2283023501","https://openalex.org/W2559655401","https://openalex.org/W2601846890","https://openalex.org/W2612690371","https://openalex.org/W2889046694","https://openalex.org/W2915116423","https://openalex.org/W2938647293","https://openalex.org/W2963417959","https://openalex.org/W2964210434","https://openalex.org/W2964265968","https://openalex.org/W2981407587","https://openalex.org/W3000307639","https://openalex.org/W3007343725","https://openalex.org/W3102037514","https://openalex.org/W3103457637","https://openalex.org/W3113425034","https://openalex.org/W3132264265","https://openalex.org/W4212774754","https://openalex.org/W4284712609"],"related_works":["https://openalex.org/W4236459141","https://openalex.org/W4205304778","https://openalex.org/W2020252434","https://openalex.org/W73248859","https://openalex.org/W2584253892","https://openalex.org/W2350324449","https://openalex.org/W1572705989","https://openalex.org/W119381072","https://openalex.org/W2034033896","https://openalex.org/W2087062149"],"abstract_inverted_index":{"Dynamical":[0],"systems":[1,49],"theory":[2],"has":[3],"recently":[4],"been":[5],"applied":[6],"in":[7,25,55],"optimization":[8],"to":[9,112,130],"prove":[10,42,128],"that":[11,119],"gradient":[12,120],"descent":[13,121],"algorithms":[14],"bypass":[15],"so-called":[16],"strict":[17],"saddle":[18,124],"points":[19,97],"of":[20,45,60,95,98],"the":[21,31,46,61,99,143],"loss":[22,102],"function.":[23],"However,":[24],"many":[26],"modern":[27],"machine":[28,70],"learning":[29,71],"applications,":[30],"required":[32],"regularity":[33,62],"conditions":[34],"are":[35],"not":[36],"satisfied.":[37],"In":[38],"this":[39],"paper,":[40],"we":[41,57,117,127],"a":[43,51,74,92],"variant":[44],"relevant":[47],"dynamical":[48],"result,":[50],"center-stable":[52],"manifold":[53],"theorem,":[54],"which":[56],"relax":[58],"some":[59],"requirements.":[63],"We":[64],"explore":[65],"its":[66],"relevance":[67],"for":[68,104],"various":[69],"tasks,":[72],"with":[73,87],"particular":[75],"focus":[76],"on":[77,91,142],"shallow":[78,105],"rectified":[79],"linear":[80],"unit":[81],"(ReLU)":[82],"and":[83,107],"leaky":[84,108],"ReLU":[85,106,109],"networks":[86,110],"scalar":[88],"input.":[89],"Building":[90],"detailed":[93],"examination":[94],"critical":[96],"square":[100],"integral":[101],"function":[103],"relative":[111],"an":[113,139],"affine":[114],"target":[115],"function,":[116],"show":[118],"circumvents":[122],"most":[123],"points.":[125],"Furthermore,":[126],"convergence":[129],"global":[131],"minima":[132],"under":[133],"favourable":[134],"initialization":[135],"conditions,":[136],"quantified":[137],"by":[138],"explicit":[140],"threshold":[141],"limiting":[144],"loss.":[145]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-27T14:29:43.386196","created_date":"2025-10-10T00:00:00"}