{"id":"https://openalex.org/W7223293282","doi":"https://doi.org/10.48550/arxiv.2610.30729","title":"Width and Depth Limits Commute During Training in Linear Residual Networks","display_name":"Width and Depth Limits Commute During Training in Linear Residual Networks","publication_year":2026,"publication_date":"2026-10-22","ids":{"openalex":"https://openalex.org/W7223293282","doi":"https://doi.org/10.48550/arxiv.2610.30729"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.30729","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.30729","pdf_url":"https://arxiv.org/pdf/2610.30729","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.30729","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5159318951","display_name":"Ruijia Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhang, Ruijia","raw_affiliation_strings":["Department of Applied Mathematics and Statistics Johns Hopkins University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Applied Mathematics and Statistics Johns Hopkins University","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089509060","display_name":"Soufiane Hayou","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hayou, Soufiane","raw_affiliation_strings":["Department of Applied Mathematics and Statistics Johns Hopkins University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Applied Mathematics and Statistics Johns Hopkins University","institution_ids":["https://openalex.org/I145311948"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I145311948"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.7318999767303467,"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":0.7318999767303467,"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/T10320","display_name":"Neural Networks and Applications","score":0.21299999952316284,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.010300000198185444,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/infinite-width-limit","display_name":"infinite-width limit","score":0.7409999966621399},{"id":"https://openalex.org/keywords/gradient-descent-learning","display_name":"gradient descent learning","score":0.45500001311302185},{"id":"https://openalex.org/keywords/gradient-clipping","display_name":"gradient clipping","score":0.9300000071525574},{"id":"https://openalex.org/keywords/mean-square-convergence","display_name":"mean square convergence","score":0.6549999713897705},{"id":"https://openalex.org/keywords/convergence-in-probability","display_name":"convergence in probability","score":0.9110000133514404},{"id":"https://openalex.org/keywords/neural-network-initialization","display_name":"neural network initialization","score":0.36399999260902405}],"concepts":[{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.7943999767303467},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.7294999957084656},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.7055000066757202},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.642300009727478},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.6284000277519226},{"id":"https://openalex.org/C7321624","wikidata":"https://www.wikidata.org/wiki/Q205","display_name":"Infinity","level":2,"score":0.6068000197410583},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.6051999926567078},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5685999989509583},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5123000144958496},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.5077000260353088},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.45500001311302185},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.41600000858306885},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.3930000066757202},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.38420000672340393},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.29580000042915344},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.2944999933242798},{"id":"https://openalex.org/C2776799497","wikidata":"https://www.wikidata.org/wiki/Q484298","display_name":"Surface (topology)","level":2,"score":0.2858999967575073},{"id":"https://openalex.org/C61445026","wikidata":"https://www.wikidata.org/wiki/Q217608","display_name":"Fixed point","level":2,"score":0.2842000126838684},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.27970001101493835},{"id":"https://openalex.org/C19448537","wikidata":"https://www.wikidata.org/wiki/Q33540","display_name":"Limit of a function","level":3,"score":0.2782999873161316},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.2766000032424927},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2522999942302704},{"id":"https://openalex.org/C95763700","wikidata":"https://www.wikidata.org/wiki/Q578985","display_name":"Convergence of random variables","level":3,"score":0.25220000743865967}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.30729","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.30729","pdf_url":"https://arxiv.org/pdf/2610.30729","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},{"id":"doi:10.48550/arxiv.2610.30729","doi":"https://doi.org/10.48550/arxiv.2610.30729","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.30729","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.30729","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.30729","pdf_url":"https://arxiv.org/pdf/2610.30729","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7223293282.pdf","grobid_xml":"https://content.openalex.org/works/W7223293282.grobid-xml"},"referenced_works_count":8,"referenced_works":["https://openalex.org/W2610971674","https://openalex.org/W2766678531","https://openalex.org/W3001279689","https://openalex.org/W3110116503","https://openalex.org/W4387226075","https://openalex.org/W4414595198","https://openalex.org/W4416438542","https://openalex.org/W7139915297"],"related_works":["https://openalex.org/W4319049797","https://openalex.org/W4414595198","https://openalex.org/W4224028035","https://openalex.org/W7139915297","https://openalex.org/W3172612647","https://openalex.org/W2995553068","https://openalex.org/W4302011693","https://openalex.org/W4387226075","https://openalex.org/W2985531965","https://openalex.org/W7126661406"],"abstract_inverted_index":{"Recent":[0],"work":[1],"established":[2],"that":[3],"the":[4,24,29,54,61,78,99,113,125],"infinite-width":[5,100],"and":[6,34,60,93,101],"infinite-depth":[7,102],"limits":[8,103],"of":[9,72],"residual":[10,16,51],"networks":[11],"commute":[12,104],"at":[13],"initialization":[14],"when":[15],"branches":[17],"are":[18],"scaled":[19],"by":[20],"$1/\\sqrt{depth}$.":[21],"In":[22],"particular,":[23],"network":[25,79],"output":[26,80],"converges":[27,81],"to":[28,37,47,96],"same":[30,55],"limit":[31],"whether":[32],"width":[33,63,92],"depth":[35,94],"tend":[36,95],"infinity":[38],"sequentially":[39],"or":[40],"jointly.":[41],"We":[42],"extend":[43],"this":[44],"commutativity":[45],"result":[46],"training":[48,108],"in":[49,82,120,128],"linear":[50],"networks.":[52],"Under":[53],"$1/\\sqrt{depth}$":[56],"residual-branch":[57],"scaling":[58],"(Depth-$\u03bc$P)":[59],"$\u03bc$P":[62],"parameterization,":[64],"we":[65,123],"prove":[66],"that,":[67],"for":[68],"any":[69,106],"fixed":[70],"number":[71],"gradient-descent":[73],"steps":[74],"with":[75],"gradient":[76,111],"clipping,":[77,112],"$L^2$":[83,115],"along":[84],"every":[85],"joint":[86],"width--depth":[87],"growth":[88],"path":[89],"as":[90],"both":[91],"infinity.":[97],"Consequently,":[98],"throughout":[105],"finite":[107],"horizon.":[109],"Without":[110],"corresponding":[114],"convergence":[116,127],"does":[117],"not":[118],"hold":[119],"general,":[121],"but":[122],"establish":[124],"weaker":[126],"probability.":[129]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-10-24T00:00:00"}