{"id":"https://openalex.org/W7162119809","doi":"https://doi.org/10.48550/arxiv.2605.22480","title":"Implicit Regularization of Mini-Batch Training in Graph Neural Networks","display_name":"Implicit Regularization of Mini-Batch Training in Graph Neural Networks","publication_year":2026,"publication_date":"2026-05-21","ids":{"openalex":"https://openalex.org/W7162119809","doi":"https://doi.org/10.48550/arxiv.2605.22480"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.22480","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.22480","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.22480","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136778726","display_name":"Clement Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Clement","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042223897","display_name":"Antoine Vialle","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vialle, Antoine","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082401199","display_name":"Robin Vaysse","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vaysse, Robin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5020404511","display_name":"Thomas Bonald","orcid":"https://orcid.org/0000-0003-0468-0384"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bonald, Thomas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9549999833106995,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9549999833106995,"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/T12676","display_name":"Machine Learning and ELM","score":0.008200000040233135,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.004699999932199717,"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/embedding","display_name":"Embedding","score":0.6733999848365784},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5615000128746033},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.5139999985694885},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.47450000047683716},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.46959999203681946},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.4625999927520752},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.42669999599456787},{"id":"https://openalex.org/keywords/fraction","display_name":"Fraction (chemistry)","score":0.41589999198913574},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.3912000060081482}],"concepts":[{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6733999848365784},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5615000128746033},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.5139999985694885},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5081999897956848},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.47450000047683716},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.47119998931884766},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.46959999203681946},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.4625999927520752},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.45399999618530273},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.42669999599456787},{"id":"https://openalex.org/C149629883","wikidata":"https://www.wikidata.org/wiki/Q660926","display_name":"Fraction (chemistry)","level":2,"score":0.41589999198913574},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.3912000060081482},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.38370001316070557},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38269999623298645},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.3756999969482422},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.373199999332428},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.3386000096797943},{"id":"https://openalex.org/C128115575","wikidata":"https://www.wikidata.org/wiki/Q5597083","display_name":"Graph factorization","level":5,"score":0.3303000032901764},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2989000082015991},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.2985000014305115},{"id":"https://openalex.org/C121194460","wikidata":"https://www.wikidata.org/wiki/Q856741","display_name":"Random walk","level":2,"score":0.2930999994277954},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2824000120162964},{"id":"https://openalex.org/C184720557","wikidata":"https://www.wikidata.org/wiki/Q7825049","display_name":"Topology (electrical circuits)","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C75564084","wikidata":"https://www.wikidata.org/wiki/Q5597085","display_name":"Graph embedding","level":3,"score":0.27730000019073486},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.27070000767707825},{"id":"https://openalex.org/C146380142","wikidata":"https://www.wikidata.org/wiki/Q1137726","display_name":"Directed graph","level":2,"score":0.2606000006198883},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.22480","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.22480","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.22480","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.22480","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"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":{"Mini-batch":[0],"training":[1,11,52,65],"of":[2,57,68,74,154,160],"Graph":[3],"Neural":[4],"Networks":[5],"(GNNs)":[6],"is":[7,131],"fundamentally":[8],"different":[9],"from":[10],"on":[12,53,66],"i.i.d.":[13],"data:":[14],"sampling":[15],"a":[16,72,105,113,145,158,167],"subgraph":[17,56],"alters":[18],"the":[19,44,54,75,101,109,118,134,152],"topology":[20],"and":[21,36,78,95,137,163],"introduces":[22],"boundary":[23],"effects,":[24],"leading":[25],"prior":[26],"work":[27],"to":[28,88,108,133],"develop":[29],"structure-aware":[30],"samplers":[31],"that":[32,43,97],"preserve":[33],"local":[34,123],"connectivity":[35],"reduce":[37],"embedding":[38],"variance.":[39],"Surprisingly,":[40],"we":[41,83],"demonstrate":[42],"simplest":[45],"possible":[46],"scheme,":[47],"Random":[48],"Node":[49],"Sampling":[50],"(RNS),":[51],"induced":[55],"uniformly":[58],"sampled":[59,102],"nodes,":[60],"matches":[61],"or":[62],"outperforms":[63],"full-graph":[64,135],"8":[67],"10":[69],"datasets":[70],"at":[71],"fraction":[73],"wall-clock":[76],"time":[77],"memory.":[79],"To":[80],"explain":[81],"this,":[82],"apply":[84],"backward":[85],"error":[86],"analysis":[87,150],"graph":[89,155],"mini-batch":[90,110],"Stochastic":[91],"Gradient":[92],"Descent":[93],"(SGD)":[94],"show":[96],"it":[98,125],"implicitly":[99],"minimizes":[100],"loss":[103,130],"plus":[104],"regularizer":[106],"proportional":[107],"gradient":[111],"variance,":[112,143],"quantity":[114],"directly":[115],"shaped":[116],"by":[117],"sampler.":[119],"Although":[120],"RNS":[121,165],"discards":[122],"structure,":[124],"produces":[126],"mini-batches":[127],"whose":[128,138],"expected":[129],"closer":[132],"loss,":[136],"per-batch":[139],"gradients":[140],"have":[141],"lower":[142],"yielding":[144],"better":[146],"implicit":[147,161],"objective.":[148],"Our":[149],"reframes":[151],"choice":[153],"sampler":[156],"as":[157,166],"form":[159],"regularization,":[162],"identifies":[164],"strong,":[168],"theoretically":[169],"grounded":[170],"method":[171],"for":[172],"scalable":[173],"GNN":[174],"training.":[175]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-23T00:00:00"}