{"id":"https://openalex.org/W7224802457","doi":"https://doi.org/10.48550/arxiv.2610.38773","title":"Rennala NSGD: Asynchronous Stochastic Optimization Beyond Euclidean Geometry","display_name":"Rennala NSGD: Asynchronous Stochastic Optimization Beyond Euclidean Geometry","publication_year":2026,"publication_date":"2026-10-28","ids":{"openalex":"https://openalex.org/W7224802457","doi":"https://doi.org/10.48550/arxiv.2610.38773"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.38773","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.38773","pdf_url":"https://arxiv.org/pdf/2610.38773","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","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.38773","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103273832","display_name":"Igor Sokolov","orcid":"https://orcid.org/0000-0002-4914-5943","observed_orcids":["https://orcid.org/0000-0002-4914-5943"]},"institutions":[{"id":"https://openalex.org/I71920554","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38","country_code":"SA","type":"education","lineage":["https://openalex.org/I71920554"]}],"countries":["SA"],"is_corresponding":false,"raw_author_name":"Sokolov, Igor","raw_affiliation_strings":["Center of Excellence for Generative AI , King Abdullah University of Science and Technology (KAUST) , Thuwal , Saudi Arabia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center of Excellence for Generative AI , King Abdullah University of Science and Technology (KAUST) , Thuwal , Saudi Arabia","institution_ids":["https://openalex.org/I71920554"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135847823","display_name":"Alexander Tyurin","orcid":"https://orcid.org/0000-0003-4963-8227","observed_orcids":["https://orcid.org/0000-0003-4963-8227"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tyurin, Alexander","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5036598221","display_name":"Peter Richt\u00e1rik","orcid":"https://orcid.org/0000-0003-4380-5848","observed_orcids":["https://orcid.org/0000-0003-4380-5848"]},"institutions":[{"id":"https://openalex.org/I71920554","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38","country_code":"SA","type":"education","lineage":["https://openalex.org/I71920554"]}],"countries":["SA"],"is_corresponding":false,"raw_author_name":"Richt\u00e1rik, Peter","raw_affiliation_strings":["KAUST"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAUST","institution_ids":["https://openalex.org/I71920554"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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":1.0,"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":1.0,"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/T10963","display_name":"Advanced Optimization Algorithms Research","score":0.0,"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/T10249","display_name":"Distributed Control Multi-Agent Systems","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/non-euclidean-geometry","display_name":"non-Euclidean geometry","score":0.7929999828338623},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"stochastic gradient descent","score":0.7160000205039978},{"id":"https://openalex.org/keywords/normed-spaces","display_name":"normed spaces","score":0.734000027179718},{"id":"https://openalex.org/keywords/oracle-complexity","display_name":"oracle complexity","score":0.5090000033378601},{"id":"https://openalex.org/keywords/nonconvex-optimization","display_name":"non-convex optimization","score":0.7770000100135803},{"id":"https://openalex.org/keywords/distributed-optimization","display_name":"distributed optimization","score":0.5440000295639038},{"id":"https://openalex.org/keywords/asynchronous-algorithms","display_name":"asynchronous algorithms","score":0.7039999961853027},{"id":"https://openalex.org/keywords/large-language-model-training","display_name":"large language model training","score":0.800000011920929}],"concepts":[{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.6428999900817871},{"id":"https://openalex.org/C129782007","wikidata":"https://www.wikidata.org/wiki/Q162886","display_name":"Euclidean geometry","level":2,"score":0.6169000267982483},{"id":"https://openalex.org/C39927690","wikidata":"https://www.wikidata.org/wiki/Q11197","display_name":"Logarithm","level":2,"score":0.616599977016449},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.4672999978065491},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.46459999680519104},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4627000093460083},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.4564000070095062},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.4366999864578247},{"id":"https://openalex.org/C191795146","wikidata":"https://www.wikidata.org/wiki/Q3878446","display_name":"Norm (philosophy)","level":2,"score":0.42820000648498535},{"id":"https://openalex.org/C186450821","wikidata":"https://www.wikidata.org/wiki/Q17295","display_name":"Euclidean space","level":2,"score":0.4246000051498413},{"id":"https://openalex.org/C110202963","wikidata":"https://www.wikidata.org/wiki/Q1783542","display_name":"Convex geometry","level":5,"score":0.38850000500679016},{"id":"https://openalex.org/C2777027219","wikidata":"https://www.wikidata.org/wiki/Q1284190","display_name":"Constant (computer programming)","level":2,"score":0.36959999799728394},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.36649999022483826},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.35350000858306885},{"id":"https://openalex.org/C157972887","wikidata":"https://www.wikidata.org/wiki/Q463359","display_name":"Convex optimization","level":3,"score":0.353300005197525},{"id":"https://openalex.org/C18015164","wikidata":"https://www.wikidata.org/wiki/Q6935000","display_name":"Multiplicative noise","level":5,"score":0.34389999508857727},{"id":"https://openalex.org/C10784197","wikidata":"https://www.wikidata.org/wiki/Q233858","display_name":"Non-Euclidean geometry","level":3,"score":0.3418999910354614},{"id":"https://openalex.org/C134912446","wikidata":"https://www.wikidata.org/wiki/Q5166516","display_name":"Convex body","level":4,"score":0.3361000120639801},{"id":"https://openalex.org/C194387892","wikidata":"https://www.wikidata.org/wiki/Q1747770","display_name":"Stochastic optimization","level":2,"score":0.3336000144481659},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.33000001311302185},{"id":"https://openalex.org/C120174047","wikidata":"https://www.wikidata.org/wiki/Q847073","display_name":"Euclidean distance","level":2,"score":0.3296000063419342},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.3257000148296356},{"id":"https://openalex.org/C2776637919","wikidata":"https://www.wikidata.org/wiki/Q624380","display_name":"Descent (aeronautics)","level":2,"score":0.3149999976158142},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.3005000054836273},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.2793999910354614},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.2773999869823456},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2750999927520752},{"id":"https://openalex.org/C112680207","wikidata":"https://www.wikidata.org/wiki/Q714886","display_name":"Regular polygon","level":2,"score":0.2669999897480011},{"id":"https://openalex.org/C190470478","wikidata":"https://www.wikidata.org/wiki/Q2370229","display_name":"Invariant (physics)","level":2,"score":0.265500009059906},{"id":"https://openalex.org/C145446738","wikidata":"https://www.wikidata.org/wiki/Q319913","display_name":"Convex function","level":3,"score":0.2574000060558319},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.25270000100135803}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.38773","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.38773","pdf_url":"https://arxiv.org/pdf/2610.38773","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.2610.38773","doi":"https://doi.org/10.48550/arxiv.2610.38773","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.38773","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:2610.38773","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.38773","pdf_url":"https://arxiv.org/pdf/2610.38773","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":[{"id":"https://openalex.org/F4320322320","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224802457.pdf","grobid_xml":"https://content.openalex.org/works/W7224802457.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4416330759","https://openalex.org/W4416255819","https://openalex.org/W4417252840","https://openalex.org/W7164858199","https://openalex.org/W2347281465","https://openalex.org/W4414809339","https://openalex.org/W778657980","https://openalex.org/W7117126534","https://openalex.org/W7106804492","https://openalex.org/W2750685566"],"abstract_inverted_index":{"The":[0,100],"empirical":[1],"success":[2],"of":[3,19,37,74,113,196],"matrix-aware":[4],"optimizers,":[5],"such":[6],"as":[7],"Muon":[8],"(Jordan":[9],"et":[10],"al.,":[11],"2024)":[12],"and":[13,137,173],"Shampoo":[14],"(arXiv:1802.09568),":[15],"highlights":[16],"the":[17,34,72,92,97,103,109,114,118,124,186,193],"potential":[18],"non-Euclidean":[20,135],"geometry":[21,106],"for":[22,57],"deep":[23],"learning,":[24],"including":[25],"large":[26],"language":[27],"model":[28],"(LLM)":[29],"training.":[30],"Despite":[31],"this":[32,163],"progress,":[33],"theoretical":[35],"understanding":[36],"stochastic":[38,53],"optimization":[39,60],"in":[40,47,91],"general":[41],"normed":[42,76],"spaces":[43,77],"remains":[44],"limited,":[45],"particularly":[46],"distributed":[48,171],"settings.":[49],"We":[50],"analyze":[51],"minibatch":[52,126],"gradient":[54],"descent":[55],"(SGD)":[56],"smooth":[58],"nonconvex":[59],"with":[61],"respect":[62],"to":[63,130,157,201],"an":[64],"arbitrary":[65],"norm":[66,94],"on":[67,105,162],"a":[68,86,148,153,158,169],"finite-dimensional":[69],"space.":[70,116],"Using":[71],"framework":[73],"$\u03ba$-regular":[75],"(arXiv:0809.0813),":[78],"we":[79,165],"derive":[80],"high-probability":[81],"oracle":[82],"complexity":[83,177],"bounds":[84,101,144],"under":[85,179],"light-tail":[87],"condition":[88],"stated":[89],"directly":[90],"dual":[93,115],"induced":[95],"by":[96,146],"update":[98],"norm.":[99],"make":[102],"dependence":[104,156,195],"explicit":[107],"through":[108],"regularity":[110],"constant":[111],"$\u03ba$":[112],"In":[117,185],"Euclidean":[119,149,187],"case,":[120],"our":[121,190],"bound":[122,191],"recovers":[123],"standard":[125],"SGD":[127,199],"rate":[128],"up":[129,200],"logarithmic":[131,159,202],"factors.":[132,203],"For":[133],"some":[134],"geometries":[136],"noise":[138,150],"models,":[139],"it":[140],"is":[141],"sharper":[142],"than":[143],"obtained":[145],"converting":[147],"assumption,":[151],"reducing":[152],"linear":[154],"dimension":[155],"one.":[160],"Building":[161],"framework,":[164],"introduce":[166],"Rennala":[167,198],"NSGD,":[168],"semi-asynchronous":[170],"algorithm,":[172],"establish":[174],"wall-clock":[175],"time":[176],"guarantees":[178],"fixed,":[180],"heterogeneous":[181],"worker":[182],"computation":[183],"times.":[184],"case":[188],"($\u03ba=1$),":[189],"matches":[192],"parameter":[194],"classical":[197]},"counts_by_year":[],"updated_date":"2026-11-02T09:45:03.842371","created_date":"2026-10-31T00:00:00"}