{"id":"https://openalex.org/W3158301457","doi":"https://doi.org/10.1109/ciss50987.2021.9400243","title":"Connection of Diagonal Hessian Estimates to Natural Gradients in Stochastic Optimization","display_name":"Connection of Diagonal Hessian Estimates to Natural Gradients in Stochastic Optimization","publication_year":2021,"publication_date":"2021-03-24","ids":{"openalex":"https://openalex.org/W3158301457","doi":"https://doi.org/10.1109/ciss50987.2021.9400243","mag":"3158301457"},"language":"en","primary_location":{"id":"doi:10.1109/ciss50987.2021.9400243","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ciss50987.2021.9400243","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5068293468","display_name":"Shiqing Sun","orcid":null},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shiqing Sun","raw_affiliation_strings":["The Johns Hopkins University, Baltimore, MD"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Johns Hopkins University, Baltimore, MD","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066890566","display_name":"James C. Spall","orcid":"https://orcid.org/0000-0001-8034-8864"},"institutions":[{"id":"https://openalex.org/I2802946424","display_name":"Johns Hopkins University Applied Physics Laboratory","ror":"https://ror.org/029pp9z10","country_code":"US","type":"facility","lineage":["https://openalex.org/I145311948","https://openalex.org/I2802946424"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"James C. Spall","raw_affiliation_strings":["The Johns Hopkins University, Applied Physics Lab, Baltimore, MD"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Johns Hopkins University, Applied Physics Lab, Baltimore, MD","institution_ids":["https://openalex.org/I2802946424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0434522,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"65","issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9998999834060669,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9998999834060669,"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":0.9994000196456909,"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/T10320","display_name":"Neural Networks and Applications","score":0.9968000054359436,"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/hessian-matrix","display_name":"Hessian matrix","score":0.9640034437179565},{"id":"https://openalex.org/keywords/diagonal","display_name":"Diagonal","score":0.7123738527297974},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.5508402585983276},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5208616852760315},{"id":"https://openalex.org/keywords/stochastic-approximation","display_name":"Stochastic approximation","score":0.49121755361557007},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.4578001797199249},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.4571017026901245},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.45305106043815613},{"id":"https://openalex.org/keywords/fisher-information","display_name":"Fisher information","score":0.42889973521232605},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.35300010442733765},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.32850441336631775},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.23008465766906738},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.19700956344604492},{"id":"https://openalex.org/keywords/geometry","display_name":"Geometry","score":0.12204301357269287}],"concepts":[{"id":"https://openalex.org/C203616005","wikidata":"https://www.wikidata.org/wiki/Q620495","display_name":"Hessian matrix","level":2,"score":0.9640034437179565},{"id":"https://openalex.org/C130367717","wikidata":"https://www.wikidata.org/wiki/Q189791","display_name":"Diagonal","level":2,"score":0.7123738527297974},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.5508402585983276},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5208616852760315},{"id":"https://openalex.org/C55479107","wikidata":"https://www.wikidata.org/wiki/Q97663916","display_name":"Stochastic approximation","level":3,"score":0.49121755361557007},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4578001797199249},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.4571017026901245},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.45305106043815613},{"id":"https://openalex.org/C29406490","wikidata":"https://www.wikidata.org/wiki/Q1420659","display_name":"Fisher information","level":2,"score":0.42889973521232605},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35300010442733765},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32850441336631775},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.23008465766906738},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.19700956344604492},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.12204301357269287},{"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/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ciss50987.2021.9400243","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ciss50987.2021.9400243","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W19621276","https://openalex.org/W1522301498","https://openalex.org/W1607198972","https://openalex.org/W1844261860","https://openalex.org/W1970789124","https://openalex.org/W2065699608","https://openalex.org/W2097115711","https://openalex.org/W2146502635","https://openalex.org/W2194775991","https://openalex.org/W2474694648","https://openalex.org/W2523246573","https://openalex.org/W2565654137","https://openalex.org/W2886919371","https://openalex.org/W2907225497","https://openalex.org/W2949935872","https://openalex.org/W2964121744","https://openalex.org/W2964309400","https://openalex.org/W2973164215","https://openalex.org/W3011817532","https://openalex.org/W3086499488","https://openalex.org/W3140968660","https://openalex.org/W6600828528","https://openalex.org/W6681435938","https://openalex.org/W6754055986","https://openalex.org/W6757751764"],"related_works":["https://openalex.org/W2924468784","https://openalex.org/W3143650729","https://openalex.org/W3177326532","https://openalex.org/W2042173174","https://openalex.org/W359561291","https://openalex.org/W4399282524","https://openalex.org/W1994333428","https://openalex.org/W4390895039","https://openalex.org/W4286899070","https://openalex.org/W4323366756"],"abstract_inverted_index":{"With":[0],"massive":[1,24],"resurgence":[2],"of":[3,15,52,67,87,93,143,155,205],"artificial":[4],"intelligence,":[5],"statistical":[6,32],"learning":[7,27,176],"theory":[8],"and":[9,43,106,112,168,187],"information":[10,75],"science,":[11],"the":[12,50,64,68,72,90,94,100,104,109,113,122,141,152,156,170,174,182],"core":[13],"technology":[14],"AI,":[16],"are":[17,29,45,81],"getting":[18],"growing":[19],"attention.":[20],"To":[21],"deal":[22],"with":[23,71,202],"data,":[25],"efficient":[26],"algorithms":[28],"required":[30],"in":[31,140,178,185,188,198],"learning.":[33,88,145],"In":[34,125,208],"deep":[35,144],"learning,":[36],"natural":[37,78,189],"gradient":[38,79,190],"algorithms,":[39],"such":[40],"as":[41],"AdaGrad":[42],"Adam,":[44],"widely":[46],"used,":[47],"motivated":[48],"by":[49,99],"idea":[51],"Newton's":[53],"approach":[54],"that":[55,194],"applies":[56,169],"second-order":[57,65,148],"derivatives":[58],"to":[59,83,215],"rescale":[60],"gradients.":[61],"By":[62,180],"approximating":[63],"geometry":[66],"empirical":[69,73,95,110],"loss":[70,96,200],"Fisher":[74],"matrix":[76,158],"(FIM),":[77],"methods":[80],"expected":[82,123],"obtain":[84],"extra":[85],"efficiency":[86],"However,":[89],"exact":[91],"curvature":[92,201],"is":[97],"described":[98],"Hessian":[101,114,137,157,206],"matrix,":[102],"not":[103],"FIM,":[105],"biases":[107],"between":[108],"FIM":[111],"always":[115],"exist":[116],"before":[117],"convergence,":[118],"which":[119],"will":[120],"affect":[121],"efficiency.":[124],"this":[126],"paper,":[127],"we":[128,192,211],"present":[129],"a":[130,147,213],"new":[131],"stochastic":[132,138,165],"optimization":[133],"algorithm,":[134,149],"diagSG":[135,150,186,195],"(diagonal":[136],"gradient),":[139],"setting":[142],"As":[146],"estimates":[151],"diagonal":[153,171],"entries":[154,172],"at":[159],"each":[160],"iteration":[161],"through":[162],"simultaneous":[163],"perturbation":[164],"approximation":[166,204],"(SPSA)":[167],"for":[173],"adaptive":[175],"rate":[177],"optimization.":[179],"comparing":[181],"rescaling":[183],"matrices":[184],"methods,":[191],"argue":[193],"possess":[196],"advantages":[197],"characterizing":[199],"better":[203],"diagonals.":[207],"practical":[209],"part,":[210],"provide":[212],"experiment":[214],"endorse":[216],"our":[217],"argument.":[218]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}