{"id":"https://openalex.org/W4368754957","doi":"https://doi.org/10.48550/arxiv.2305.01868","title":"Pre-train and Search: Efficient Embedding Table Sharding with Pre-trained Neural Cost Models","display_name":"Pre-train and Search: Efficient Embedding Table Sharding with Pre-trained Neural Cost Models","publication_year":2023,"publication_date":"2023-05-03","ids":{"openalex":"https://openalex.org/W4368754957","doi":"https://doi.org/10.48550/arxiv.2305.01868"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2305.01868","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.01868","pdf_url":"https://arxiv.org/pdf/2305.01868","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,"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"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2305.01868","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5058071176","display_name":"Daochen Zha","orcid":"https://orcid.org/0000-0002-6677-7504"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zha, Daochen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100808706","display_name":"Louis Feng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feng, Louis","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101672911","display_name":"Liang Luo","orcid":"https://orcid.org/0009-0009-4403-5604"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079877271","display_name":"Bhargav Bhushanam","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bhushanam, Bhargav","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101916543","display_name":"Zirui Liu","orcid":"https://orcid.org/0000-0001-9062-6565"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Zirui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102510826","display_name":"Yusuo Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Yusuo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004002767","display_name":"Jade Nie","orcid":"https://orcid.org/0009-0008-1664-2805"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nie, Jade","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034586871","display_name":"Yuzhen Huang","orcid":"https://orcid.org/0000-0002-9536-7918"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Yuzhen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106668475","display_name":"Yuandong Tian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tian, Yuandong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047988079","display_name":"Arun Kejariwal","orcid":"https://orcid.org/0009-0006-6172-2973"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kejariwal, Arun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5068477431","display_name":"Xia Hu","orcid":"https://orcid.org/0000-0003-2234-3226"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Xia","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":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/T10036","display_name":"Advanced Neural Network Applications","score":0.996399998664856,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.996399998664856,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9807000160217285,"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/T10320","display_name":"Neural Networks and Applications","score":0.9609000086784363,"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/computer-science","display_name":"Computer science","score":0.8087841272354126},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.788989782333374},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.734309196472168},{"id":"https://openalex.org/keywords/table","display_name":"Table (database)","score":0.6896341443061829},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.45078244805336},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.43046462535858154},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4265420436859131},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.34182924032211304},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.18224141001701355},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.06386867165565491}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8087841272354126},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.788989782333374},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.734309196472168},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.6896341443061829},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.45078244805336},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.43046462535858154},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4265420436859131},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34182924032211304},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.18224141001701355},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.06386867165565491},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2305.01868","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.01868","pdf_url":"https://arxiv.org/pdf/2305.01868","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,"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.2305.01868","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2305.01868","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,"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":"pmh:oai:arXiv.org:2305.01868","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.01868","pdf_url":"https://arxiv.org/pdf/2305.01868","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,"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":[{"display_name":"Industry, innovation and infrastructure","id":"https://metadata.un.org/sdg/9","score":0.49000000953674316}],"awards":[{"id":"https://openalex.org/G8272762023","display_name":"CAREER: Human-Centric Big Network Embedding","funder_award_id":"2224843","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320310133","display_name":"Rice University","ror":"https://ror.org/008zs3103"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4368754957.pdf","grobid_xml":"https://content.openalex.org/works/W4368754957.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2378211422","https://openalex.org/W2745001401","https://openalex.org/W4321353415","https://openalex.org/W2130974462","https://openalex.org/W2028665553","https://openalex.org/W2086519370","https://openalex.org/W972276598","https://openalex.org/W2087343574","https://openalex.org/W4246352526","https://openalex.org/W2049082574"],"abstract_inverted_index":{"Sharding":[0],"a":[1,39,52],"large":[2],"machine":[3],"learning":[4,103],"model":[5],"across":[6],"multiple":[7],"devices":[8],"to":[9,50,58,86,122,162,188],"balance":[10],"the":[11,27,60,64,88,130,153,156,184,199],"costs":[12,28,61,182],"is":[13,19,23,32,49],"important":[14],"in":[15,101,167,180,204],"distributed":[16],"training.":[17],"This":[18],"challenging":[20],"because":[21],"partitioning":[22],"NP-hard,":[24],"and":[25,30,41,54,107,133,140,150,201],"estimating":[26],"accurately":[29],"efficiently":[31],"difficult.":[33],"In":[34],"this":[35,76,99],"work,":[36],"we":[37,80,208],"explore":[38],"\"pre-train,":[40,200],"search\"":[42,202],"paradigm":[43,203],"for":[44,110,206],"efficient":[45,71],"sharding.":[46,113],"The":[47],"idea":[48,100],"pre-train":[51],"universal":[53],"once-for-all":[55],"neural":[56,116],"network":[57],"predict":[59],"of":[62,198],"all":[63],"possible":[65],"shards,":[66],"which":[67,186],"serves":[68],"as":[69],"an":[70,83,168],"sharding":[72,90,95,125,135,158],"simulator.":[73],"Built":[74],"upon":[75],"pre-trained":[77],"cost":[78,117],"model,":[79],"then":[81],"perform":[82],"online":[84],"search":[85,139],"identify":[87],"best":[89,131],"plans":[91,136],"given":[92],"any":[93],"specific":[94],"task.":[96],"We":[97],"instantiate":[98],"deep":[102],"recommendation":[104],"models":[105,118],"(DLRMs)":[106],"propose":[108],"NeuroShard":[109,114,148,176],"embedding":[111,174,181],"table":[112],"pre-trains":[115],"on":[119,155],"augmented":[120],"tables":[121],"cover":[123],"various":[124],"scenarios.":[126],"Then":[127],"it":[128],"identifies":[129],"column-wise":[132],"table-wise":[134],"with":[137,172],"beam":[138],"greedy":[141],"grid":[142],"search,":[143],"respectively.":[144],"Experiments":[145],"show":[146],"that":[147],"significantly":[149],"consistently":[151],"outperforms":[152],"state-of-the-art":[154],"benchmark":[157],"dataset,":[159],"achieving":[160],"up":[161],"23.8%":[163],"improvement.":[164,193],"When":[165],"deployed":[166],"ultra-large":[169],"production":[170],"DLRM":[171],"multi-terabyte":[173],"tables,":[175],"achieves":[177],"11.6%":[178],"improvement":[179],"over":[183],"state-of-the-art,":[185],"translates":[187],"6.6%":[189],"end-to-end":[190],"training":[191],"throughput":[192],"To":[194],"facilitate":[195],"future":[196],"research":[197],"ML":[205],"Systems,":[207],"open-source":[209],"our":[210],"code":[211],"at":[212],"https://github.com/daochenzha/neuroshard":[213]},"counts_by_year":[],"updated_date":"2026-08-26T07:47:46.906454","created_date":"2023-05-05T00:00:00"}