{"id":"https://openalex.org/W4407949476","doi":"https://doi.org/10.1109/cdc56724.2024.10886208","title":"A Stochastic Quasi-Newton Method for Non-convex Optimization with Non-uniform Smoothness","display_name":"A Stochastic Quasi-Newton Method for Non-convex Optimization with Non-uniform Smoothness","publication_year":2024,"publication_date":"2024-12-16","ids":{"openalex":"https://openalex.org/W4407949476","doi":"https://doi.org/10.1109/cdc56724.2024.10886208"},"language":"en","primary_location":{"id":"doi:10.1109/cdc56724.2024.10886208","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cdc56724.2024.10886208","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 63rd Conference on Decision and Control (CDC)","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/A5045330094","display_name":"Zhenyu Sun","orcid":"https://orcid.org/0000-0001-5902-2691"},"institutions":[{"id":"https://openalex.org/I111979921","display_name":"Northwestern University","ror":"https://ror.org/000e0be47","country_code":"US","type":"education","lineage":["https://openalex.org/I111979921"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhenyu Sun","raw_affiliation_strings":["Northwestern University,Electrical & Computer Engineering,Evanston,USA,IL 60208"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northwestern University,Electrical & Computer Engineering,Evanston,USA,IL 60208","institution_ids":["https://openalex.org/I111979921"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5085511405","display_name":"Ermin Wei","orcid":"https://orcid.org/0000-0002-8035-484X"},"institutions":[{"id":"https://openalex.org/I111979921","display_name":"Northwestern University","ror":"https://ror.org/000e0be47","country_code":"US","type":"education","lineage":["https://openalex.org/I111979921"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ermin Wei","raw_affiliation_strings":["Northwestern University,Electrical & Computer Engineering and Industrial Engineering & Management Science,Evanston,USA,IL 60208"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northwestern University,Electrical & Computer Engineering and Industrial Engineering & Management Science,Evanston,USA,IL 60208","institution_ids":["https://openalex.org/I111979921"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I111979921"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"5668","last_page":"5674"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10963","display_name":"Advanced Optimization Algorithms Research","score":0.9842000007629395,"subfield":{"id":"https://openalex.org/subfields/2612","display_name":"Numerical Analysis"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10963","display_name":"Advanced Optimization Algorithms Research","score":0.9842000007629395,"subfield":{"id":"https://openalex.org/subfields/2612","display_name":"Numerical Analysis"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9812999963760376,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9502000212669373,"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/smoothness","display_name":"Smoothness","score":0.7659778594970703},{"id":"https://openalex.org/keywords/regular-polygon","display_name":"Regular polygon","score":0.5355261564254761},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.5231469869613647},{"id":"https://openalex.org/keywords/newtons-method","display_name":"Newton's method","score":0.5124902725219727},{"id":"https://openalex.org/keywords/convex-optimization","display_name":"Convex optimization","score":0.48306018114089966},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.45070430636405945},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.4408086836338043},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.41549187898635864},{"id":"https://openalex.org/keywords/mathematical-analysis","display_name":"Mathematical analysis","score":0.22833505272865295},{"id":"https://openalex.org/keywords/geometry","display_name":"Geometry","score":0.13156333565711975},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.10653534531593323},{"id":"https://openalex.org/keywords/nonlinear-system","display_name":"Nonlinear system","score":0.08225688338279724}],"concepts":[{"id":"https://openalex.org/C102634674","wikidata":"https://www.wikidata.org/wiki/Q868473","display_name":"Smoothness","level":2,"score":0.7659778594970703},{"id":"https://openalex.org/C112680207","wikidata":"https://www.wikidata.org/wiki/Q714886","display_name":"Regular polygon","level":2,"score":0.5355261564254761},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.5231469869613647},{"id":"https://openalex.org/C85189116","wikidata":"https://www.wikidata.org/wiki/Q374195","display_name":"Newton's method","level":3,"score":0.5124902725219727},{"id":"https://openalex.org/C157972887","wikidata":"https://www.wikidata.org/wiki/Q463359","display_name":"Convex optimization","level":3,"score":0.48306018114089966},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.45070430636405945},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.4408086836338043},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.41549187898635864},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.22833505272865295},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.13156333565711975},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.10653534531593323},{"id":"https://openalex.org/C158622935","wikidata":"https://www.wikidata.org/wiki/Q660848","display_name":"Nonlinear system","level":2,"score":0.08225688338279724},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cdc56724.2024.10886208","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cdc56724.2024.10886208","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 63rd Conference on Decision and Control (CDC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":37,"referenced_works":["https://openalex.org/W1987083649","https://openalex.org/W2042173174","https://openalex.org/W2051434435","https://openalex.org/W2079705627","https://openalex.org/W2963470657","https://openalex.org/W2963941964","https://openalex.org/W2964303576","https://openalex.org/W2970986510","https://openalex.org/W2993258424","https://openalex.org/W2997241048","https://openalex.org/W3131853808","https://openalex.org/W3134948707","https://openalex.org/W4250589301","https://openalex.org/W4293775970","https://openalex.org/W6629379589","https://openalex.org/W6629536081","https://openalex.org/W6640275870","https://openalex.org/W6676105031","https://openalex.org/W6698038680","https://openalex.org/W6698102009","https://openalex.org/W6734509502","https://openalex.org/W6737778391","https://openalex.org/W6738155353","https://openalex.org/W6740708716","https://openalex.org/W6743395824","https://openalex.org/W6752458581","https://openalex.org/W6753122105","https://openalex.org/W6759119893","https://openalex.org/W6766675349","https://openalex.org/W6772129282","https://openalex.org/W6781810711","https://openalex.org/W6791321639","https://openalex.org/W6794588271","https://openalex.org/W6794728472","https://openalex.org/W6838914693","https://openalex.org/W6850378916","https://openalex.org/W6863952952"],"related_works":["https://openalex.org/W2393022482","https://openalex.org/W2377346130","https://openalex.org/W2361092061","https://openalex.org/W2319775965","https://openalex.org/W2357314690","https://openalex.org/W2191886813","https://openalex.org/W1986317414","https://openalex.org/W2795976185","https://openalex.org/W2163394011","https://openalex.org/W2353606679"],"abstract_inverted_index":{"Classical":[0],"convergence":[1,174],"analyses":[2],"for":[3,116,138],"optimization":[4],"algorithms":[5,103],"rely":[6],"on":[7],"the":[8,28,35,50,56,60,67,111,124,167,189],"widely-adopted":[9],"uniform":[10],"smoothness":[11,29,51,87],"assumption.":[12],"However,":[13],"recent":[14],"experimental":[15],"studies":[16,125],"have":[17],"demonstrated":[18],"that":[19,49,81,184],"many":[20],"machine":[21],"learning":[22],"problems":[23],"exhibit":[24],"nonuniform":[25],"smoothness,":[26,96],"meaning":[27],"factor":[30],"is":[31,72],"a":[32,40,73,146],"function":[33],"of":[34,39,94,126],"model":[36],"parameter":[37],"instead":[38],"universal":[41],"constant.":[42],"In":[43],"particular,":[44],"it":[45],"has":[46,99],"been":[47],"observed":[48],"grows":[52],"with":[53,176],"respect":[54],"to":[55,78,109],"gradient":[57,89,106,158],"norm":[58],"along":[59],"training":[61],"trajectory.":[62],"Motivated":[63],"by":[64,104],"this":[65,92,142],"phenomenon,":[66],"recently":[68],"introduced":[69],"($L_{0},":[70],"L_{1}$)smoothness":[71],"more":[74,136],"general":[75],"notion,":[76],"compared":[77],"traditional":[79],"L-smoothness,":[80],"captures":[82],"such":[83],"positive":[84],"relationship":[85],"between":[86],"and":[88,135,160,172],"norm.":[90],"Under":[91],"type":[93],"non-uniform":[95],"existing":[97],"literature":[98],"designed":[100],"stochastic":[101,148],"first-order":[102,120],"utilizing":[105],"clipping":[107,159],"techniques":[108],"obtain":[110],"optimal":[112],"$\\mathcal{O}\\left(\\epsilon^{-3}\\right)$":[113,169],"sample":[114,170],"complexity":[115,171],"finding":[117],"an":[118],"$\\epsilon$-approximate":[119],"stationary":[121],"solution.":[122],"Nevertheless,":[123],"quasiNewton":[127],"methods":[128],"are":[129],"still":[130],"lacking.":[131],"Considering":[132],"higher":[133],"accuracy":[134],"robustness":[137],"quasi-Newton":[139,149],"methods,":[140],"in":[141,155],"paper":[143],"we":[144],"propose":[145],"fast":[147],"method":[150],"when":[151],"there":[152],"exists":[153],"non-uniformity":[154],"smoothness.":[156],"Leveraging":[157],"variance":[161],"reduction,":[162],"our":[163,185],"algorithm":[164,187],"can":[165],"achieve":[166],"best-known":[168],"enjoys":[173],"speedup":[175],"simple":[177],"hyperparameter":[178],"tuning.":[179],"Our":[180],"numerical":[181],"experiments":[182],"show":[183],"proposed":[186],"outperforms":[188],"state-of-the-art":[190],"approaches.":[191]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}