{"id":"https://openalex.org/W7221809970","doi":"https://doi.org/10.48550/arxiv.2610.14344","title":"On the Depth Transfer of Neural Optimizers via L-Smoothness","display_name":"On the Depth Transfer of Neural Optimizers via L-Smoothness","publication_year":2026,"publication_date":"2026-10-09","ids":{"openalex":"https://openalex.org/W7221809970","doi":"https://doi.org/10.48550/arxiv.2610.14344"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.14344","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.14344","pdf_url":"https://arxiv.org/pdf/2610.14344","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/2610.14344","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5060378047","display_name":"Maximilian Wong","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I40347166","display_name":"University of Chicago","ror":"https://ror.org/024mw5h28","country_code":"US","type":"education","lineage":["https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wong, Maximilian","raw_affiliation_strings":["University of Chicago"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chicago","institution_ids":["https://openalex.org/I40347166"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5158754852","display_name":"Chenyu Xu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu, Chenyu","raw_affiliation_strings":["Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5158719908","display_name":"Jiajin Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I141945490","display_name":"University of British Columbia","ror":"https://ror.org/03rmrcq20","country_code":"CA","type":"education","lineage":["https://openalex.org/I141945490"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Li, Jiajin","raw_affiliation_strings":["Sauder School of Business University of British Columbia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sauder School of Business University of British Columbia","institution_ids":["https://openalex.org/I141945490"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5158694058","display_name":"Yiping Lu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lu, Yiping","raw_affiliation_strings":["Beijing International Center for Mathematical Research Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing International Center for Mathematical Research Peking University","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"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":0.9961000084877014,"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"}},"topics":[{"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"}},{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.001500000013038516,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0007999999797903001,"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/hyperparameter-transfer","display_name":"hyperparameter transfer","score":0.8629999756813049},{"id":"https://openalex.org/keywords/lipschitz-constant","display_name":"Lipschitz constant","score":0.8209999799728394},{"id":"https://openalex.org/keywords/learning-rate-scaling","display_name":"learning rate scaling","score":0.7639999985694885},{"id":"https://openalex.org/keywords/adamw","display_name":"AdamW","score":0.7670000195503235},{"id":"https://openalex.org/keywords/muon-optimizer","display_name":"Muon optimizer","score":0.9269999861717224},{"id":"https://openalex.org/keywords/training-dynamics","display_name":"training dynamics","score":0.43700000643730164},{"id":"https://openalex.org/keywords/gpt-2","display_name":"GPT-2","score":0.7089999914169312},{"id":"https://openalex.org/keywords/resnet","display_name":"ResNet","score":0.8579999804496765}],"concepts":[{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.7315999865531921},{"id":"https://openalex.org/C22324862","wikidata":"https://www.wikidata.org/wiki/Q652707","display_name":"Lipschitz continuity","level":2,"score":0.6567999720573425},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.6216999888420105},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5942999720573425},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.59170001745224},{"id":"https://openalex.org/C102634674","wikidata":"https://www.wikidata.org/wiki/Q868473","display_name":"Smoothness","level":2,"score":0.5278000235557556},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.51910001039505},{"id":"https://openalex.org/C195065555","wikidata":"https://www.wikidata.org/wiki/Q214881","display_name":"Curvature","level":2,"score":0.45249998569488525},{"id":"https://openalex.org/C2777027219","wikidata":"https://www.wikidata.org/wiki/Q1284190","display_name":"Constant (computer programming)","level":2,"score":0.43529999256134033},{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.42739999294281006},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4124999940395355},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.40310001373291016},{"id":"https://openalex.org/C202615002","wikidata":"https://www.wikidata.org/wiki/Q783507","display_name":"Differentiable function","level":2,"score":0.4027999937534332},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.39410001039505005},{"id":"https://openalex.org/C39927690","wikidata":"https://www.wikidata.org/wiki/Q11197","display_name":"Logarithm","level":2,"score":0.3873000144958496},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.382999986410141},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3758000135421753},{"id":"https://openalex.org/C81299745","wikidata":"https://www.wikidata.org/wiki/Q334269","display_name":"Transfer function","level":2,"score":0.367900013923645},{"id":"https://openalex.org/C165464430","wikidata":"https://www.wikidata.org/wiki/Q1570441","display_name":"Parameterized complexity","level":2,"score":0.3319000005722046},{"id":"https://openalex.org/C2776175482","wikidata":"https://www.wikidata.org/wiki/Q1195816","display_name":"Transfer (computing)","level":2,"score":0.3280999958515167},{"id":"https://openalex.org/C2779668893","wikidata":"https://www.wikidata.org/wiki/Q2250422","display_name":"Constant curvature","level":3,"score":0.3127000033855438},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.31049999594688416},{"id":"https://openalex.org/C90377204","wikidata":"https://www.wikidata.org/wiki/Q1052594","display_name":"Uniform boundedness","level":3,"score":0.30070000886917114},{"id":"https://openalex.org/C132459708","wikidata":"https://www.wikidata.org/wiki/Q744069","display_name":"Extrapolation","level":2,"score":0.29829999804496765},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.295199990272522},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.28949999809265137},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.28610000014305115},{"id":"https://openalex.org/C2780388253","wikidata":"https://www.wikidata.org/wiki/Q5421508","display_name":"Exponent","level":2,"score":0.2547999918460846},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.25459998846054077},{"id":"https://openalex.org/C130367717","wikidata":"https://www.wikidata.org/wiki/Q189791","display_name":"Diagonal","level":2,"score":0.2533000111579895}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.14344","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.14344","pdf_url":"https://arxiv.org/pdf/2610.14344","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.2610.14344","doi":"https://doi.org/10.48550/arxiv.2610.14344","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.14344","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.14344","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.14344","pdf_url":"https://arxiv.org/pdf/2610.14344","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/W7221809970.pdf","grobid_xml":"https://content.openalex.org/works/W7221809970.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7128488623","https://openalex.org/W7134925487","https://openalex.org/W4417143336","https://openalex.org/W4387226075","https://openalex.org/W4416291078","https://openalex.org/W7162036416","https://openalex.org/W4415878270","https://openalex.org/W7147008202","https://openalex.org/W7214768121","https://openalex.org/W7165146692"],"abstract_inverted_index":{"Scaling":[0],"deep":[1],"learning":[2],"models":[3],"efficiently":[4],"requires":[5],"predictable":[6],"performance":[7],"transfer":[8,15,44,75,239],"across":[9,248],"compute":[10],"regimes.":[11],"Establishing":[12],"reliable":[13],"hyperparameter":[14,43,74],"allows":[16],"optimal":[17,246],"training":[18],"dynamics":[19],"and":[20,41,51,60,81,116,134,154,206,217,225,234],"structural":[21],"choices":[22],"to":[23,102,140,209],"be":[24],"discovered":[25],"on":[26,54,64,126,159,221],"low-cost":[27],"proxy":[28],"models,":[29],"avoiding":[30],"catastrophic":[31],"failures":[32],"during":[33],"massive-scale":[34],"training.":[35],"Existing":[36],"depth-transfer":[37],"analyses":[38],"are":[39],"architecture-specific":[40],"heuristic:":[42],"is":[45,100],"only":[46,130],"justified":[47],"through":[48,131],"toy":[49],"blocks,":[50],"both":[52,104],"rely":[53],"infinite-depth":[55],"dynamics,":[56],"a":[57,181,199,244,249],"specific":[58],"initialization,":[59],"feature-learning":[61],"criteria.":[62],"Building":[63],"$L$-smoothness":[65,82],"framework":[66],"of":[67,84,146,202],"Xu":[68],"et":[69],"al.":[70],"(2026),":[71],"we":[72],"understand":[73],"as":[76,152],"how":[77],"do":[78],"the":[79,85,89,112,123,137,167,171,175,194,210],"Lipschitz":[80,133],"constants":[83,105],"loss":[86],"scale":[87,183],"with":[88,214,243],"depth":[90,148],"$K$?":[91],"We":[92],"show":[93],"that":[94,232],"inverse-depth":[95,203],"residual":[96,128,142,204,229],"branch":[97,129],"scaling":[98,205],"$\u03b1=\u0398(1/K)$":[99],"sufficient":[101],"make":[103],"depth-independent":[106],"at":[107],"fixed":[108],"width,":[109],"uniformly":[110],"over":[111],"bounded":[113,192],"parameter":[114],"domain":[115],"for":[117,190],"arbitrary":[118],"$p\\to":[119],"q$":[120],"geometries.":[121],"Because":[122],"bounds":[124,179],"depend":[125],"each":[127],"its":[132,207],"smoothness":[135,168],"constants,":[136],"argument":[138],"applies":[139],"general":[141],"networks":[143],"(including":[144],"branches":[145],"block":[147],"$\\geq":[149],"2$":[150],"such":[151],"transformer":[153],"ResNet":[155],"blocks)":[156,230],"without":[157,184],"assumptions":[158],"initialization":[160],"or":[161],"correlations":[162],"between":[163],"layer":[164],"weights.":[165],"Since":[166],"constant":[169],"controls":[170],"curvature":[172,188],"penalty":[173],"in":[174,193,252],"descent":[176],"inequality,":[177],"our":[178],"motivate":[180],"learning-rate":[182,238],"an":[185],"additional":[186],"depth-dependent":[187],"correction":[189],"updates":[191],"prescribed":[195],"norm.":[196],"This":[197],"provides":[198],"geometric":[200],"explanation":[201],"application":[208],"operator-norm":[211],"geometries":[212],"associated":[213],"AdamW,":[215],"Muon,":[216],"MOGA.":[218],"Learning-rate":[219],"sweeps":[220],"GPT-2":[222],"(depths":[223],"2--128)":[224],"CIFAR-100":[226],"ResNets":[227],"(8--32":[228],"confirm":[231],"Muon":[233],"MOGA":[235],"exhibit":[236],"stable":[237],"under":[240],"these":[241],"prescriptions,":[242],"common":[245],"rate":[247],"64-fold":[250],"increase":[251],"depth.":[253]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-10-15T00:00:00"}