{"id":"https://openalex.org/W4214572086","doi":"https://doi.org/10.1109/tai.2022.3153593","title":"Projective Fisher Information for Natural Gradient Descent","display_name":"Projective Fisher Information for Natural Gradient Descent","publication_year":2022,"publication_date":"2022-02-25","ids":{"openalex":"https://openalex.org/W4214572086","doi":"https://doi.org/10.1109/tai.2022.3153593"},"language":"en","primary_location":{"id":"doi:10.1109/tai.2022.3153593","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tai.2022.3153593","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5044545345","display_name":"Piyush Kaul","orcid":"https://orcid.org/0000-0003-3117-2377"},"institutions":[{"id":"https://openalex.org/I68891433","display_name":"Indian Institute of Technology Delhi","ror":"https://ror.org/049tgcd06","country_code":"IN","type":"education","lineage":["https://openalex.org/I68891433"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Piyush Kaul","raw_affiliation_strings":["Department of Electrical Engineering, Indian Institute of Technology - Delhi, New Delhi, India"],"raw_orcid":"https://orcid.org/0000-0003-3117-2377","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, Indian Institute of Technology - Delhi, New Delhi, India","institution_ids":["https://openalex.org/I68891433"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066116024","display_name":"Brejesh Lall","orcid":"https://orcid.org/0000-0003-2677-3071"},"institutions":[{"id":"https://openalex.org/I68891433","display_name":"Indian Institute of Technology Delhi","ror":"https://ror.org/049tgcd06","country_code":"IN","type":"education","lineage":["https://openalex.org/I68891433"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Brejesh Lall","raw_affiliation_strings":["Department of Electrical Engineering, Indian Institute of Technology - Delhi, New Delhi, India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, Indian Institute of Technology - Delhi, New Delhi, India","institution_ids":["https://openalex.org/I68891433"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I68891433"],"apc_list":null,"apc_paid":null,"fwci":0.3936,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.64690952,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":"4","issue":"2","first_page":"304","last_page":"314"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9993000030517578,"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/T10320","display_name":"Neural Networks and Applications","score":0.9993000030517578,"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/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.9965000152587891,"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":0.996399998664856,"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/gradient-descent","display_name":"Gradient descent","score":0.7141094207763672},{"id":"https://openalex.org/keywords/fisher-information","display_name":"Fisher information","score":0.6969780325889587},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6283884644508362},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5351734161376953},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.5092154741287231},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5048225522041321},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.47786110639572144},{"id":"https://openalex.org/keywords/projection","display_name":"Projection (relational algebra)","score":0.4431923031806946},{"id":"https://openalex.org/keywords/kronecker-delta","display_name":"Kronecker delta","score":0.4318754971027374},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.42373162508010864},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.3779551684856415},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.30176717042922974},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2882745563983917}],"concepts":[{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.7141094207763672},{"id":"https://openalex.org/C29406490","wikidata":"https://www.wikidata.org/wiki/Q1420659","display_name":"Fisher information","level":2,"score":0.6969780325889587},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6283884644508362},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5351734161376953},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.5092154741287231},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5048225522041321},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.47786110639572144},{"id":"https://openalex.org/C57493831","wikidata":"https://www.wikidata.org/wiki/Q3134666","display_name":"Projection (relational algebra)","level":2,"score":0.4431923031806946},{"id":"https://openalex.org/C39482219","wikidata":"https://www.wikidata.org/wiki/Q192826","display_name":"Kronecker delta","level":2,"score":0.4318754971027374},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42373162508010864},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3779551684856415},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.30176717042922974},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2882745563983917},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tai.2022.3153593","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tai.2022.3153593","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":53,"referenced_works":["https://openalex.org/W41554520","https://openalex.org/W44815768","https://openalex.org/W639708223","https://openalex.org/W1496461287","https://openalex.org/W1536252645","https://openalex.org/W1949248870","https://openalex.org/W1970789124","https://openalex.org/W2064675550","https://openalex.org/W2091696547","https://openalex.org/W2099111195","https://openalex.org/W2100495367","https://openalex.org/W2112796928","https://openalex.org/W2121927366","https://openalex.org/W2194775991","https://openalex.org/W2474694648","https://openalex.org/W2476548250","https://openalex.org/W2744316982","https://openalex.org/W2750384547","https://openalex.org/W2888674590","https://openalex.org/W2919115771","https://openalex.org/W2953571781","https://openalex.org/W2963433607","https://openalex.org/W2967791890","https://openalex.org/W3003789643","https://openalex.org/W3014847672","https://openalex.org/W3026534984","https://openalex.org/W3034736335","https://openalex.org/W3035182906","https://openalex.org/W3037045207","https://openalex.org/W3118608800","https://openalex.org/W3132174721","https://openalex.org/W3132846247","https://openalex.org/W3157235145","https://openalex.org/W3175206419","https://openalex.org/W4229633200","https://openalex.org/W4246193833","https://openalex.org/W6629536081","https://openalex.org/W6683107984","https://openalex.org/W6684191040","https://openalex.org/W6684497637","https://openalex.org/W6687566353","https://openalex.org/W6693969136","https://openalex.org/W6721263324","https://openalex.org/W6739901393","https://openalex.org/W6743688258","https://openalex.org/W6752294818","https://openalex.org/W6753753558","https://openalex.org/W6762718338","https://openalex.org/W6773468611","https://openalex.org/W6779861650","https://openalex.org/W6787972765","https://openalex.org/W6790818818","https://openalex.org/W6791183400"],"related_works":["https://openalex.org/W2895097035","https://openalex.org/W4206903459","https://openalex.org/W2754816816","https://openalex.org/W4366280654","https://openalex.org/W3160167280","https://openalex.org/W4362706668","https://openalex.org/W4231621013","https://openalex.org/W3171021120","https://openalex.org/W3008318776","https://openalex.org/W2041416246"],"abstract_inverted_index":{"Improvements":[0],"in":[1],"neural":[2,64,93,128,204],"network":[3],"optimization":[4,189,202],"algorithms":[5],"have":[6,96],"enabled":[7],"shorter":[8],"training":[9],"times":[10],"and":[11,39,60,81,103,126,166,173],"the":[12,36,40,58,77,88,106,115,132,147,183,188],"ability":[13],"to":[14,56,76,79,203],"reach":[15],"state-of-the-art":[16],"performance":[17,42],"on":[18,146],"various":[19],"machine":[20,47],"learning":[21,48],"tasks.":[22],"Fisher":[23,50,116,153],"information":[24,51,117,154],"based":[25,145],"natural":[26,67,141,197],"gradient":[27,68,142,198],"descent":[28,69,143,199],"is":[29,70,86,162],"one":[30],"such":[31],"second-order":[32],"method":[33,74],"that":[34,176],"improves":[35],"convergence":[37],"speed":[38],"final":[41],"metric":[43],"achieved":[44],"for":[45,104,118,201],"many":[46],"algorithms.":[49],"matrices":[52],"are":[53],"also":[54,136],"helpful":[55],"analyze":[57],"properties":[59],"expected":[61],"behavior":[62],"of":[63,101,120,123,149,152,159,210],"networks.":[65],"However,":[66],"a":[71,97,138,207],"high":[72,99],"complexity":[73,140],"due":[75],"need":[78],"maintain":[80],"invert":[82],"covariance":[83],"matrices.":[84],"This":[85],"especially":[87],"case":[89],"with":[90,156,206],"modern":[91],"deep":[92],"networks,":[94],"which":[95,105,161],"very":[98],"number":[100,209],"parameters,":[102,211],"problem":[107],"often":[108],"becomes":[109],"computationally":[110,163],"unfeasible.":[111],"We":[112,135,169],"suggest":[113],"using":[114],"analysis":[119,172],"parameter":[121],"space":[122],"fully":[124],"connected":[125],"convolutional":[127],"networks":[129,205],"without":[130],"calculating":[131],"matrix":[133],"itself.":[134],"propose":[137],"lower":[139],"algorithm":[144],"projection":[148],"Kronecker":[150],"factors":[151],"combined":[155],"recursive":[157],"calculation":[158],"inverses,":[160],"less":[164],"complex":[165],"more":[167],"stable.":[168],"finally":[170],"share":[171],"results":[174],"showing":[175],"all":[177],"these":[178],"optimizations":[179],"do":[180],"not":[181],"impact":[182],"accuracy":[184],"while":[185],"considerably":[186],"lowering":[187],"process\u2019s":[190],"complexity.":[191],"These":[192],"improvements":[193],"should":[194],"enable":[195],"applying":[196],"methods":[200],"larger":[208],"than":[212],"possible":[213],"previously.":[214]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}