{"id":"https://openalex.org/W3081233814","doi":"https://doi.org/10.1145/3394486.3403265","title":"Rich Information is Affordable: A Systematic Performance Analysis of Second-order Optimization Using K-FAC","display_name":"Rich Information is Affordable: A Systematic Performance Analysis of Second-order Optimization Using K-FAC","publication_year":2020,"publication_date":"2020-08-20","ids":{"openalex":"https://openalex.org/W3081233814","doi":"https://doi.org/10.1145/3394486.3403265","mag":"3081233814"},"language":"en","primary_location":{"id":"doi:10.1145/3394486.3403265","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3394486.3403265","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3394486.3403265","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3394486.3403265","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101655156","display_name":"Yuichiro Ueno","orcid":"https://orcid.org/0000-0001-8763-2075"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Yuichiro Ueno","raw_affiliation_strings":["Tokyo Institute of Technology & AIST-Tokyo Tech RWBC-OIL, AIST, Tokyo, Japan"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology & AIST-Tokyo Tech RWBC-OIL, AIST, Tokyo, Japan","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076237237","display_name":"Kazuki Osawa","orcid":"https://orcid.org/0000-0001-6390-9797"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Kazuki Osawa","raw_affiliation_strings":["Tokyo Institute of Technology, Tokyo, Japan"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology, Tokyo, Japan","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040483976","display_name":"Yohei Tsuji","orcid":"https://orcid.org/0000-0001-8108-2324"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Yohei Tsuji","raw_affiliation_strings":["Tokyo Institute of Technology, Tokyo, Japan"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology, Tokyo, Japan","institution_ids":["https://openalex.org/I114531698"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050692022","display_name":"Akira Naruse","orcid":"https://orcid.org/0000-0002-3140-0854"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Akira Naruse","raw_affiliation_strings":["NVIDIA, Tokyo, Japan"],"affiliations":[{"raw_affiliation_string":"NVIDIA, Tokyo, Japan","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024747717","display_name":"Rio Yokota","orcid":"https://orcid.org/0000-0001-7573-7873"},"institutions":[{"id":"https://openalex.org/I114531698","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960","country_code":"JP","type":"education","lineage":["https://openalex.org/I114531698"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Rio Yokota","raw_affiliation_strings":["Tokyo Institute of Technology & AIST-Tokyo Tech RWBC-OIL, AIST, Tokyo, Japan"],"affiliations":[{"raw_affiliation_string":"Tokyo Institute of Technology & AIST-Tokyo Tech RWBC-OIL, AIST, Tokyo, Japan","institution_ids":["https://openalex.org/I114531698"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5101655156"],"corresponding_institution_ids":["https://openalex.org/I114531698"],"apc_list":null,"apc_paid":null,"fwci":1.2172,"has_fulltext":true,"cited_by_count":9,"citation_normalized_percentile":{"value":0.79599948,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"2145","last_page":"2153"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11447","display_name":"Blind Source Separation Techniques","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T11447","display_name":"Blind Source Separation Techniques","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9994000196456909,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12676","display_name":"Machine Learning and ELM","score":0.9986000061035156,"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":0.7809895873069763},{"id":"https://openalex.org/keywords/hyperparameter","display_name":"Hyperparameter","score":0.7579241991043091},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.6620502471923828},{"id":"https://openalex.org/keywords/fisher-information","display_name":"Fisher information","score":0.5640914440155029},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5351522564888},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5201440453529358},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4770716428756714},{"id":"https://openalex.org/keywords/preconditioner","display_name":"Preconditioner","score":0.4758843183517456},{"id":"https://openalex.org/keywords/hyperparameter-optimization","display_name":"Hyperparameter optimization","score":0.4723706841468811},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4719861149787903},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3842350244522095},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.38195863366127014},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.34190666675567627},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3249431252479553},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.14891162514686584},{"id":"https://openalex.org/keywords/iterative-method","display_name":"Iterative method","score":0.1364762783050537},{"id":"https://openalex.org/keywords/support-vector-machine","display_name":"Support vector machine","score":0.12210077047348022}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7809895873069763},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.7579241991043091},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.6620502471923828},{"id":"https://openalex.org/C29406490","wikidata":"https://www.wikidata.org/wiki/Q1420659","display_name":"Fisher information","level":2,"score":0.5640914440155029},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5351522564888},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5201440453529358},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4770716428756714},{"id":"https://openalex.org/C167431342","wikidata":"https://www.wikidata.org/wiki/Q1754327","display_name":"Preconditioner","level":3,"score":0.4758843183517456},{"id":"https://openalex.org/C10485038","wikidata":"https://www.wikidata.org/wiki/Q48996162","display_name":"Hyperparameter optimization","level":3,"score":0.4723706841468811},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4719861149787903},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3842350244522095},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.38195863366127014},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.34190666675567627},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3249431252479553},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.14891162514686584},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.1364762783050537},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.12210077047348022},{"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/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3394486.3403265","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3394486.3403265","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3394486.3403265","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3394486.3403265","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3394486.3403265","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3394486.3403265","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3282004645","display_name":null,"funder_award_id":"JPMJCR","funder_id":"https://openalex.org/F4320338075","funder_display_name":"Core Research for Evolutional Science and Technology"},{"id":"https://openalex.org/G3612518180","display_name":null,"funder_award_id":"JPMJCR19F5","funder_id":"https://openalex.org/F4320338075","funder_display_name":"Core Research for Evolutional Science and Technology"},{"id":"https://openalex.org/G8769122518","display_name":null,"funder_award_id":"JPMJCR19F5","funder_id":"https://openalex.org/F4320334789","funder_display_name":"Japan Science and Technology Agency"}],"funders":[{"id":"https://openalex.org/F4320311508","display_name":"National Institute of Advanced Industrial Science and Technology","ror":"https://ror.org/01703db54"},{"id":"https://openalex.org/F4320322704","display_name":"Tokyo Institute of Technology","ror":"https://ror.org/0112mx960"},{"id":"https://openalex.org/F4320334789","display_name":"Japan Science and Technology Agency","ror":"https://ror.org/00097mb19"},{"id":"https://openalex.org/F4320338075","display_name":"Core Research for Evolutional Science and Technology","ror":"https://ror.org/00097mb19"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3081233814.pdf","grobid_xml":"https://content.openalex.org/works/W3081233814.grobid-xml"},"referenced_works_count":13,"referenced_works":["https://openalex.org/W1970789124","https://openalex.org/W2057332538","https://openalex.org/W2098086734","https://openalex.org/W2117539524","https://openalex.org/W2769856846","https://openalex.org/W2900824371","https://openalex.org/W2903376718","https://openalex.org/W2913668833","https://openalex.org/W2955454939","https://openalex.org/W2962747323","https://openalex.org/W2962781217","https://openalex.org/W2962883024","https://openalex.org/W2967791890"],"related_works":["https://openalex.org/W4381737452","https://openalex.org/W4281646320","https://openalex.org/W3014750173","https://openalex.org/W4287818966","https://openalex.org/W3139241485","https://openalex.org/W4323894855","https://openalex.org/W4295309597","https://openalex.org/W4283697347","https://openalex.org/W4298144215","https://openalex.org/W3199608561"],"abstract_inverted_index":{"Rich":[0],"information":[1,24,59,72,93],"matrices":[2,25,60,73],"from":[3],"first":[4,152],"and":[5,15,29,45,101],"second-order":[6],"derivatives":[7],"have":[8],"many":[9],"potential":[10],"applications":[11],"in":[12,18,133],"both":[13],"theoretical":[14],"practical":[16],"problems":[17,40],"deep":[19,48],"learning.":[20],"However,":[21],"computing":[22,90],"these":[23],"is":[26,33,61,150],"extremely":[27],"expensive":[28],"this":[30,81],"enormous":[31],"cost":[32,115],"currently":[34],"limiting":[35],"its":[36],"application":[37],"to":[38,75,109,140,154,165],"important":[39],"regarding":[41],"generalization,":[42],"hyperparameter":[43],"tuning,":[44],"optimization":[46],"of":[47,52,58,97,116,137,168],"neural":[49],"networks.":[50],"One":[51],"the":[53,68,71,91,104,110,114,126,135,144,151,162],"most":[54],"challenging":[55],"use":[56,63],"cases":[57],"their":[62],"as":[64,113],"a":[65,85,118],"preconditioner":[66],"for":[67],"optimizers,":[69],"since":[70],"need":[74],"be":[76,107],"updated":[77],"every":[78],"step.":[79,121],"In":[80],"work,":[82],"we":[83],"conduct":[84],"step-by-step":[86],"performance":[87],"analysis":[88],"when":[89],"Fisher":[92,128],"matrix":[94],"during":[95],"training":[96,163],"ResNet-50":[98],"on":[99],"ImageNet,":[100],"show":[102,124],"that":[103,125,167],"overhead":[105],"can":[106,131],"reduced":[108],"same":[111,145],"amount":[112],"performing":[117],"single":[119],"SGD":[120],"We":[122],"also":[123],"resulting":[127],"preconditioned":[129],"optimizer":[130],"converge":[132],"1/3":[134],"number":[136],"epochs":[138],"compared":[139],"SGD,":[141],"while":[142,160],"achieving":[143],"Top-1":[146],"validation":[147],"accuracy.":[148],"This":[149],"work":[153],"achieve":[155],"such":[156],"accuracy":[157],"with":[158],"K-FAC":[159],"reducing":[161],"time":[164],"match":[166],"SGD.":[169]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":4}],"updated_date":"2026-04-10T15:06:20.359241","created_date":"2025-10-10T00:00:00"}