{"id":"https://openalex.org/W7165662014","doi":"https://doi.org/10.48550/arxiv.2606.22917","title":"GRAIN: Group Aggregation via Min-Norm Objective","display_name":"GRAIN: Group Aggregation via Min-Norm Objective","publication_year":2026,"publication_date":"2026-06-22","ids":{"openalex":"https://openalex.org/W7165662014","doi":"https://doi.org/10.48550/arxiv.2606.22917"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.22917","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.22917","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.22917","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134995991","display_name":"Nghia Bui","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bui, Nghia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139176213","display_name":"Jiarui Yao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yao, Jiarui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139179355","display_name":"Lijing Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Lijing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"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":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":0.6223999857902527,"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":0.6223999857902527,"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":0.1873999983072281,"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"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.06930000334978104,"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":0.6718999743461609},{"id":"https://openalex.org/keywords/traverse","display_name":"Traverse","score":0.5418999791145325},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.532800018787384},{"id":"https://openalex.org/keywords/upper-and-lower-bounds","display_name":"Upper and lower bounds","score":0.4975000023841858},{"id":"https://openalex.org/keywords/variance","display_name":"Variance (accounting)","score":0.4903999865055084},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.42559999227523804},{"id":"https://openalex.org/keywords/variance-reduction","display_name":"Variance reduction","score":0.4253000020980835},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.385699987411499},{"id":"https://openalex.org/keywords/optimization-problem","display_name":"Optimization problem","score":0.367000013589859},{"id":"https://openalex.org/keywords/suite","display_name":"Suite","score":0.36340001225471497}],"concepts":[{"id":"https://openalex.org/C102634674","wikidata":"https://www.wikidata.org/wiki/Q868473","display_name":"Smoothness","level":2,"score":0.6718999743461609},{"id":"https://openalex.org/C176809094","wikidata":"https://www.wikidata.org/wiki/Q15401496","display_name":"Traverse","level":2,"score":0.5418999791145325},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.532800018787384},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.49880000948905945},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.4975000023841858},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.4903999865055084},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4821999967098236},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.45509999990463257},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.42559999227523804},{"id":"https://openalex.org/C62644790","wikidata":"https://www.wikidata.org/wiki/Q3454689","display_name":"Variance reduction","level":3,"score":0.4253000020980835},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.41769999265670776},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.385699987411499},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.367000013589859},{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.36340001225471497},{"id":"https://openalex.org/C112680207","wikidata":"https://www.wikidata.org/wiki/Q714886","display_name":"Regular polygon","level":2,"score":0.36039999127388},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.35760000348091125},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.33329999446868896},{"id":"https://openalex.org/C82578977","wikidata":"https://www.wikidata.org/wiki/Q16773055","display_name":"Data aggregator","level":3,"score":0.328000009059906},{"id":"https://openalex.org/C157972887","wikidata":"https://www.wikidata.org/wiki/Q463359","display_name":"Convex optimization","level":3,"score":0.32440000772476196},{"id":"https://openalex.org/C2781311116","wikidata":"https://www.wikidata.org/wiki/Q83306","display_name":"Group (periodic table)","level":2,"score":0.3089999854564667},{"id":"https://openalex.org/C21080849","wikidata":"https://www.wikidata.org/wiki/Q13611879","display_name":"Data point","level":2,"score":0.30730000138282776},{"id":"https://openalex.org/C145446738","wikidata":"https://www.wikidata.org/wiki/Q319913","display_name":"Convex function","level":3,"score":0.2944999933242798},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2838999927043915},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.2831999957561493},{"id":"https://openalex.org/C22679943","wikidata":"https://www.wikidata.org/wiki/Q159375","display_name":"Standard deviation","level":2,"score":0.27810001373291016},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2718000113964081},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.27140000462532043},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.26969999074935913},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2678000032901764},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.2676999866962433},{"id":"https://openalex.org/C93693863","wikidata":"https://www.wikidata.org/wiki/Q897659","display_name":"Branch and bound","level":2,"score":0.26589998602867126},{"id":"https://openalex.org/C3020493868","wikidata":"https://www.wikidata.org/wiki/Q55631277","display_name":"Real world data","level":2,"score":0.26159998774528503},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.25929999351501465},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.22917","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.22917","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.22917","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.22917","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Learning":[0],"instability":[1],"is":[2,11,69,73],"a":[3,92,110,113,122,167,202,215],"long-standing":[4],"problem":[5,64],"across":[6,106,181],"machine":[7],"learning,":[8],"but":[9],"it":[10],"especially":[12],"acute":[13],"in":[14,102,193,198],"the":[15,63,98,127,155,162,175],"overparameterized":[16],"regime":[17],"that":[18,96],"defines":[19],"modern":[20],"deep":[21],"learning:":[22],"large":[23,58],"models":[24,60],"fine-tuned":[25],"or":[26,211],"trained":[27],"on":[28],"limited":[29],"data":[30,43,72],"traverse":[31],"flat":[32],"loss":[33],"landscapes":[34],"with":[35,112,207],"many":[36],"nearly-equivalent":[37],"minima,":[38],"and":[39,76,108,130,136,140,152,184,196],"stochastic":[40],"factors":[41],"(initialization,":[42],"order,":[44],"dropout,":[45],"hardware":[46],"non-determinism)":[47],"can":[48],"route":[49],"optimization":[50,104],"to":[51,147],"very":[52],"different":[53],"solutions.":[54],"The":[55],"rise":[56],"of":[57,117,205],"pretrained":[59],"(LPMs)":[61],"makes":[62],"more":[65],"urgent:":[66],"training":[67,94],"cost":[68,213],"high,":[70],"downstream":[71],"often":[74],"small,":[75],"repeated":[77],"runs":[78],"for":[79,170,178],"variance":[80,200],"reduction":[81],"are":[82],"prohibitive.":[83],"We":[84],"introduce":[85],"\\textbf{GRAIN}":[86],"(\\textbf{G}roup":[87],"\\textbf{A}ggregation":[88],"via":[89],"m\\textbf{IN}-norm":[90],"objective),":[91],"lightweight":[93],"algorithm":[95],"replaces":[97],"mean":[99,194],"aggregation":[100],"used":[101],"mini-batch":[103],"(both":[105],"mini-batches":[107],"within":[109],"mini-batch)":[111],"min-norm":[114,156],"convex":[115],"combination":[116],"group-wise":[118],"gradients.":[119],"\\mName":[120,171,189],"guarantees":[121],"non-negative":[123],"inner":[124],"product":[125],"between":[126],"aggregated":[128],"update":[129],"every":[131],"group":[132],"gradient,":[133],"resolving":[134],"intra-":[135],"inner-batch":[137],"gradient":[138],"conflict,":[139],"retains":[141],"an":[142],"$\\mathcal{O}(1/T)$":[143],"convergence":[144],"rate":[145],"comparable":[146],"SGD.":[148,179],"Under":[149],"mild":[150],"smoothness":[151],"absolute-continuity":[153],"assumptions,":[154],"solution":[157],"differs":[158],"almost":[159],"surely":[160],"from":[161],"arithmetic":[163],"mean,":[164],"which":[165],"yields":[166],"uniform-stability":[168],"bound":[169,177],"strictly":[172],"tighter":[173],"than":[174],"standard":[176],"Empirically":[180],"generation,":[182],"classification,":[183],"regression":[185],"at":[186],"LPM":[187],"scale,":[188],"delivers":[190],"consistent":[191],"improvements":[192],"performance":[195],"reductions":[197],"run-to-run":[199],"over":[201],"broad":[203],"suite":[204],"tasks,":[206],"no":[208],"extra":[209],"training-time":[210],"storage":[212],"beyond":[214],"single":[216],"backward":[217],"pass.":[218]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-24T00:00:00"}