{"id":"https://openalex.org/W2799070580","doi":"https://doi.org/10.1145/3184558.3193135","title":"Training 100,000 Classes on a Single Titan X in 7 Hours or 15 Minutes with 25 Titan Xs","display_name":"Training 100,000 Classes on a Single Titan X in 7 Hours or 15 Minutes with 25 Titan Xs","publication_year":2018,"publication_date":"2018-01-01","ids":{"openalex":"https://openalex.org/W2799070580","doi":"https://doi.org/10.1145/3184558.3193135","mag":"2799070580"},"language":"en","primary_location":{"id":"doi:10.1145/3184558.3193135","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3184558.3193135","pdf_url":"http://dl.acm.org/ft_gateway.cfm?id=3193135&type=pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Companion of the The Web Conference 2018 on The Web Conference 2018  - WWW '18","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"http://dl.acm.org/ft_gateway.cfm?id=3193135&type=pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5024993683","display_name":"Anshumali Shrivastava","orcid":"https://orcid.org/0000-0002-5042-2856"},"institutions":[{"id":"https://openalex.org/I74775410","display_name":"Rice University","ror":"https://ror.org/008zs3103","country_code":"US","type":"education","lineage":["https://openalex.org/I74775410"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Anshumali Shrivastava","raw_affiliation_strings":["Rice University, Houston, TX, USA"],"affiliations":[{"raw_affiliation_string":"Rice University, Houston, TX, USA","institution_ids":["https://openalex.org/I74775410"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5024993683"],"corresponding_institution_ids":["https://openalex.org/I74775410"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.02792265,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1895","last_page":"1895"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9991000294685364,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9991000294685364,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9983000159263611,"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":0.9947999715805054,"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":0.7215054631233215},{"id":"https://openalex.org/keywords/curse-of-dimensionality","display_name":"Curse of dimensionality","score":0.6792212724685669},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6553245782852173},{"id":"https://openalex.org/keywords/titan","display_name":"Titan (rocket family)","score":0.6292251944541931},{"id":"https://openalex.org/keywords/classifier","display_name":"Classifier (UML)","score":0.6227940917015076},{"id":"https://openalex.org/keywords/hash-function","display_name":"Hash function","score":0.6146225929260254},{"id":"https://openalex.org/keywords/mach-number","display_name":"Mach number","score":0.5961707234382629},{"id":"https://openalex.org/keywords/feature-hashing","display_name":"Feature hashing","score":0.594773530960083},{"id":"https://openalex.org/keywords/logarithm","display_name":"Logarithm","score":0.46930745244026184},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.46404924988746643},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4629894495010376},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4188573360443115},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.34916943311691284},{"id":"https://openalex.org/keywords/hash-table","display_name":"Hash table","score":0.3019631505012512},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1767508089542389}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7215054631233215},{"id":"https://openalex.org/C111030470","wikidata":"https://www.wikidata.org/wiki/Q1430460","display_name":"Curse of dimensionality","level":2,"score":0.6792212724685669},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6553245782852173},{"id":"https://openalex.org/C50805821","wikidata":"https://www.wikidata.org/wiki/Q1136670","display_name":"Titan (rocket family)","level":2,"score":0.6292251944541931},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.6227940917015076},{"id":"https://openalex.org/C99138194","wikidata":"https://www.wikidata.org/wiki/Q183427","display_name":"Hash function","level":2,"score":0.6146225929260254},{"id":"https://openalex.org/C165231844","wikidata":"https://www.wikidata.org/wiki/Q160669","display_name":"Mach number","level":2,"score":0.5961707234382629},{"id":"https://openalex.org/C133667856","wikidata":"https://www.wikidata.org/wiki/Q5439682","display_name":"Feature hashing","level":5,"score":0.594773530960083},{"id":"https://openalex.org/C39927690","wikidata":"https://www.wikidata.org/wiki/Q11197","display_name":"Logarithm","level":2,"score":0.46930745244026184},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46404924988746643},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4629894495010376},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4188573360443115},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34916943311691284},{"id":"https://openalex.org/C67388219","wikidata":"https://www.wikidata.org/wiki/Q207440","display_name":"Hash table","level":3,"score":0.3019631505012512},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1767508089542389},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C57879066","wikidata":"https://www.wikidata.org/wiki/Q41217","display_name":"Mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C138111711","wikidata":"https://www.wikidata.org/wiki/Q478351","display_name":"Double hashing","level":4,"score":0.0},{"id":"https://openalex.org/C146978453","wikidata":"https://www.wikidata.org/wiki/Q3798668","display_name":"Aerospace engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3184558.3193135","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3184558.3193135","pdf_url":"http://dl.acm.org/ft_gateway.cfm?id=3193135&type=pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Companion of the The Web Conference 2018 on The Web Conference 2018  - WWW '18","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3184558.3193135","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3184558.3193135","pdf_url":"http://dl.acm.org/ft_gateway.cfm?id=3193135&type=pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Companion of the The Web Conference 2018 on The Web Conference 2018  - WWW '18","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.6399999856948853,"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2799070580.pdf","grobid_xml":"https://content.openalex.org/works/W2799070580.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4381744218","https://openalex.org/W1835589799","https://openalex.org/W2767764284","https://openalex.org/W2059244188","https://openalex.org/W2144265691","https://openalex.org/W4211126162","https://openalex.org/W4205587245","https://openalex.org/W3087964089","https://openalex.org/W3158263601","https://openalex.org/W3192025065"],"abstract_inverted_index":{"In":[0,165],"this":[1,139,142,211],"talk,":[2],"I":[3,92],"will":[4,93],"present":[5],"Merged-Averaged":[6],"Classifiers":[7],"via":[8],"Hashing":[9],"(MACH)":[10],"for":[11,46],"K-classification":[12,52],"with":[13,54,73,86,115,127,172,198,255,284],"ultra-large":[14],"values":[15],"of":[16,77,90,98,108,131,189,208,220],"K.":[17],"Compared":[18],"to":[19,70,79,216,279],"traditional":[20],"one-vs-all":[21,150],"classifiers":[22,262],"that":[23,152,238],"require":[24],"$O(Kd)$":[25],"memory":[26,37],"and":[27,118,161],"inference":[28],"cost,":[29],"MACH":[30,48,66,167],"only":[31,39],"need":[32],"$O(d\u0142ogK)$":[33],"(d":[34],"is":[35,49,134,148,212,232],"dimensionality)":[36],"while":[38],"requiring":[40],"$O(K\u0142ogK":[41],"+":[42],"d\u0142ogK":[43],")$":[44],"operation":[45],"inference.":[47],"a":[50,74,122,149,202,225,248,272],"generic":[51],"algorithm,":[53],"provably":[55],"theoretical":[56],"guarantees,":[57],"without":[58],"any":[59],"assumption":[60],"on":[61,121,138,201,224,247,271],"the":[62,95,105,128,135,144,176,206,213,230,264],"relationship":[63],"between":[64],"classes.":[65,91],"uses":[67],"universal":[68],"hashing":[69],"reduce":[71],"classification":[72,84,129],"large":[75],"number":[76,89],"classes":[78,117],"few":[80],"(logarithmic":[81],"many)":[82],"independent":[83],"tasks":[85],"small":[87],"(constant)":[88],"show":[94,237],"first":[96,214],"quantification":[97],"discriminability-memory":[99],"tradeoff":[100],"in":[101,175,244,252,268],"multi-class":[102],"classification.":[103],"Using":[104],"simple":[106],"idea":[107],"hashing,":[109],"we":[110,184,239,259],"can":[111,168,240,260,276],"train":[112,241,261],"ODP":[113,242],"dataset":[114,194,267],"100,000":[116],"400,000":[119],"features":[120],"single":[123,203,226,249,273],"Titan":[124,227],"X":[125],"GPU,":[126],"accuracy":[130,137,171],"19.28%,":[132],"which":[133,275],"best-reported":[136],"dataset.":[140],"Before":[141],"work,":[143],"best":[145,207],"performing":[146],"baseline":[147],"classifier":[151],"requires":[153],"40":[154],"billion":[155],"parameters":[156],"(160":[157],"GB":[158],"model":[159,177],"size)":[160],"achieves":[162],"9%":[163,170],"accuracy.":[164],"contrast,":[166],"achieve":[169],"480x":[173],"reduction":[174],"size":[178,196],"(of":[179],"mere":[180],"0.3GB).":[181],"With":[182],"MACH,":[183],"also":[185],"demonstrate":[186,217],"complete":[187,218],"training":[188,219],"feature":[190],"extracted":[191],"fine-grained":[192,265],"imagenet":[193,266],"(compressed":[195],"104GB),":[197],"21,000":[199],"classes,":[200],"GPU.":[204],"To":[205],"our":[209],"knowledge,":[210],"work":[215],"these":[221],"extreme-class":[222],"datasets":[223,243],"X.":[228],"Furthermore,":[229],"algorithm":[231],"trivially":[233],"parallelizable.":[234],"Our":[235],"experiments":[236],"7":[245],"hours":[246,270],"GPU":[250,274],"or":[251],"15":[253],"minutes":[254],"25":[256],"GPUs.":[257,286],"Similarly,":[258],"over":[263,281],"24":[269],"be":[277],"reduced":[278],"little":[280],"1":[282],"hour":[283],"20":[285]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}