{"id":"https://openalex.org/W4280644337","doi":"https://doi.org/10.1007/s00521-022-07308-z","title":"A scaling calculus for the design and initialization of ReLU networks","display_name":"A scaling calculus for the design and initialization of ReLU networks","publication_year":2022,"publication_date":"2022-05-13","ids":{"openalex":"https://openalex.org/W4280644337","doi":"https://doi.org/10.1007/s00521-022-07308-z"},"language":"en","primary_location":{"id":"doi:10.1007/s00521-022-07308-z","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s00521-022-07308-z","pdf_url":"https://link.springer.com/content/pdf/10.1007/s00521-022-07308-z.pdf","source":{"id":"https://openalex.org/S147897268","display_name":"Neural Computing and Applications","issn_l":"0941-0643","issn":["0941-0643","1433-3058"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Neural Computing and Applications","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s00521-022-07308-z.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5059388072","display_name":"Aaron Defazio","orcid":"https://orcid.org/0000-0002-8764-3986"},"institutions":[{"id":"https://openalex.org/I2252078561","display_name":"Meta (Israel)","ror":"https://ror.org/02388em19","country_code":"IL","type":"company","lineage":["https://openalex.org/I2252078561","https://openalex.org/I4210114444"]},{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["IL","US"],"is_corresponding":true,"raw_author_name":"Aaron Defazio","raw_affiliation_strings":["Facebook AI Research New York, 770 Broadway, New York, NY, USA","Facebook AI Research New York, New York, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI Research New York, 770 Broadway, New York, NY, USA","institution_ids":["https://openalex.org/I4210114444","https://openalex.org/I2252078561"]},{"raw_affiliation_string":"Facebook AI Research New York, New York, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5019206666","display_name":"L\u00e9on Bottou","orcid":"https://orcid.org/0000-0002-9894-8128"},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]},{"id":"https://openalex.org/I2252078561","display_name":"Meta (Israel)","ror":"https://ror.org/02388em19","country_code":"IL","type":"company","lineage":["https://openalex.org/I2252078561","https://openalex.org/I4210114444"]}],"countries":["IL","US"],"is_corresponding":false,"raw_author_name":"L\u00e9on Bottou","raw_affiliation_strings":["Facebook AI Research New York, 770 Broadway, New York, NY, USA","Facebook AI Research New York, New York, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI Research New York, 770 Broadway, New York, NY, USA","institution_ids":["https://openalex.org/I4210114444","https://openalex.org/I2252078561"]},{"raw_affiliation_string":"Facebook AI Research New York, New York, USA","institution_ids":["https://openalex.org/I4210114444"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5059388072"],"corresponding_institution_ids":["https://openalex.org/I2252078561","https://openalex.org/I4210114444"],"apc_list":{"value":2390,"currency":"EUR","value_usd":2990},"apc_paid":{"value":2390,"currency":"EUR","value_usd":2990},"fwci":0.0,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.04988913,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":"34","issue":"17","first_page":"14807","last_page":"14821"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9991000294685364,"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":0.9991000294685364,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9980000257492065,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9969000220298767,"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/scaling","display_name":"Scaling","score":0.8758614659309387},{"id":"https://openalex.org/keywords/initialization","display_name":"Initialization","score":0.8693497776985168},{"id":"https://openalex.org/keywords/constant","display_name":"Constant (computer programming)","score":0.6974449157714844},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6938133835792542},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.6566748023033142},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5727850794792175},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.5103723406791687},{"id":"https://openalex.org/keywords/variance","display_name":"Variance (accounting)","score":0.4442650079727173},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.41036394238471985},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.3491460084915161},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.27617818117141724},{"id":"https://openalex.org/keywords/geometry","display_name":"Geometry","score":0.1302313208580017}],"concepts":[{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.8758614659309387},{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.8693497776985168},{"id":"https://openalex.org/C2777027219","wikidata":"https://www.wikidata.org/wiki/Q1284190","display_name":"Constant (computer programming)","level":2,"score":0.6974449157714844},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6938133835792542},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6566748023033142},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5727850794792175},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5103723406791687},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.4442650079727173},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.41036394238471985},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3491460084915161},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.27617818117141724},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.1302313208580017},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","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/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C121955636","wikidata":"https://www.wikidata.org/wiki/Q4116214","display_name":"Accounting","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1007/s00521-022-07308-z","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s00521-022-07308-z","pdf_url":"https://link.springer.com/content/pdf/10.1007/s00521-022-07308-z.pdf","source":{"id":"https://openalex.org/S147897268","display_name":"Neural Computing and Applications","issn_l":"0941-0643","issn":["0941-0643","1433-3058"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Neural Computing and Applications","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1007/s00521-022-07308-z","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s00521-022-07308-z","pdf_url":"https://link.springer.com/content/pdf/10.1007/s00521-022-07308-z.pdf","source":{"id":"https://openalex.org/S147897268","display_name":"Neural Computing and Applications","issn_l":"0941-0643","issn":["0941-0643","1433-3058"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Neural Computing and Applications","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320308737","display_name":"Facebook","ror":"https://ror.org/01zbnvs85"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4280644337.pdf","grobid_xml":"https://content.openalex.org/works/W4280644337.grobid-xml"},"referenced_works_count":35,"referenced_works":["https://openalex.org/W182586680","https://openalex.org/W1493526108","https://openalex.org/W1533861849","https://openalex.org/W1677182931","https://openalex.org/W1981401636","https://openalex.org/W2100659887","https://openalex.org/W2112796928","https://openalex.org/W2117539524","https://openalex.org/W2125930537","https://openalex.org/W2134557905","https://openalex.org/W2157017541","https://openalex.org/W2163605009","https://openalex.org/W2335728318","https://openalex.org/W2474694648","https://openalex.org/W2553303224","https://openalex.org/W2592916831","https://openalex.org/W2597507805","https://openalex.org/W2789210533","https://openalex.org/W2905541363","https://openalex.org/W2954960277","https://openalex.org/W2962821226","https://openalex.org/W2963966020","https://openalex.org/W2964065616","https://openalex.org/W2964350391","https://openalex.org/W2964515685","https://openalex.org/W2996687884","https://openalex.org/W2996931047","https://openalex.org/W3102528391","https://openalex.org/W3118608800","https://openalex.org/W3120740533","https://openalex.org/W3160955665","https://openalex.org/W4255949318","https://openalex.org/W4295773261","https://openalex.org/W4297803870","https://openalex.org/W4300537377"],"related_works":["https://openalex.org/W3204184292","https://openalex.org/W3176564347","https://openalex.org/W1985458517","https://openalex.org/W2355833770","https://openalex.org/W3031039437","https://openalex.org/W183202219","https://openalex.org/W3095877357","https://openalex.org/W2072565696","https://openalex.org/W4386821099","https://openalex.org/W2380964641"],"abstract_inverted_index":{"Abstract":[0],"We":[1,17,125],"propose":[2],"a":[3,7,37,65,102,133],"system":[4,106],"for":[5,10,93,108],"calculating":[6],"\u201cscaling":[8],"constant\u201d":[9],"layers":[11],"and":[12,34,80],"weights":[13,49,100],"of":[14,31,67,77,96,99,114,129,135],"neural":[15,32,103,119],"networks.":[16],"relate":[18,27],"this":[19],"scaling":[20,53,61],"constant":[21],"to":[22,28,58],"two":[23],"important":[24],"quantities":[25],"that":[26,36,39,47,73],"the":[29,45,51,71,74,78,84,94,97,109,127],"optimizability":[30],"networks,":[33,120],"argue":[35],"network":[38],"is":[40],"\u201cpreconditioned\u201d":[41],"via":[42],"scaling,":[43],"in":[44,64,101],"sense":[46],"all":[48],"have":[50],"same":[52],"constant,":[54],"will":[55],"be":[56,91],"easier":[57],"train.":[59],"This":[60],"calculus":[62],"results":[63],"number":[66],"consequences,":[68],"among":[69],"them":[70],"fact":[72],"geometric":[75],"mean":[76],"fan-in":[79],"fan-out,":[81,86],"rather":[82],"than":[83],"fan-in,":[85],"or":[87],"arithmetic":[88],"mean,":[89],"should":[90],"used":[92],"initialization":[95],"variance":[98],"network.":[104],"Our":[105],"allows":[107],"off-line":[110],"design":[111],"&":[112],"engineering":[113],"ReLU":[115],"(Rectified":[116],"Linear":[117],"Unit)":[118],"potentially":[121],"replacing":[122],"blind":[123],"experimentation.":[124],"verify":[126],"effectiveness":[128],"our":[130],"approach":[131],"on":[132],"set":[134],"benchmark":[136],"problems.":[137]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2022-05-22T00:00:00"}