{"id":"https://openalex.org/W4415428944","doi":"https://doi.org/10.3233/faia250855","title":"NITRO-D: Native Integer-Only Training of Deep Convolutional Neural Networks","display_name":"NITRO-D: Native Integer-Only Training of Deep Convolutional Neural Networks","publication_year":2025,"publication_date":"2025-10-21","ids":{"openalex":"https://openalex.org/W4415428944","doi":"https://doi.org/10.3233/faia250855"},"language":"en","primary_location":{"id":"doi:10.3233/faia250855","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia250855","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},"type":"book-chapter","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.3233/faia250855","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114367994","display_name":"Alberto Pirillo","orcid":null},"institutions":[{"id":"https://openalex.org/I93860229","display_name":"Politecnico di Milano","ror":"https://ror.org/01nffqt88","country_code":"IT","type":"education","lineage":["https://openalex.org/I93860229"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Alberto Pirillo","raw_affiliation_strings":["Department of Electronics, Information and Bioengineering, Politecnico di Milano, Via Ponzio 34/5, Milan, 20133, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronics, Information and Bioengineering, Politecnico di Milano, Via Ponzio 34/5, Milan, 20133, Italy","institution_ids":["https://openalex.org/I93860229"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103036647","display_name":"Luca Colombo","orcid":"https://orcid.org/0000-0002-0550-1627"},"institutions":[{"id":"https://openalex.org/I93860229","display_name":"Politecnico di Milano","ror":"https://ror.org/01nffqt88","country_code":"IT","type":"education","lineage":["https://openalex.org/I93860229"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Luca Colombo","raw_affiliation_strings":["Department of Electronics, Information and Bioengineering, Politecnico di Milano, Via Ponzio 34/5, Milan, 20133, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronics, Information and Bioengineering, Politecnico di Milano, Via Ponzio 34/5, Milan, 20133, Italy","institution_ids":["https://openalex.org/I93860229"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035547226","display_name":"Manuel Roveri","orcid":"https://orcid.org/0000-0001-7828-7687"},"institutions":[{"id":"https://openalex.org/I93860229","display_name":"Politecnico di Milano","ror":"https://ror.org/01nffqt88","country_code":"IT","type":"education","lineage":["https://openalex.org/I93860229"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Manuel Roveri","raw_affiliation_strings":["Department of Electronics, Information and Bioengineering, Politecnico di Milano, Via Ponzio 34/5, Milan, 20133, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronics, Information and Bioengineering, Politecnico di Milano, Via Ponzio 34/5, Milan, 20133, Italy","institution_ids":["https://openalex.org/I93860229"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I93860229"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.44279145,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9010000228881836,"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":0.9010000228881836,"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/quantization","display_name":"Quantization (signal processing)","score":0.6477000117301941},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5788999795913696},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5242000222206116},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5084999799728394},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.49900001287460327},{"id":"https://openalex.org/keywords/perceptron","display_name":"Perceptron","score":0.4072999954223633},{"id":"https://openalex.org/keywords/classifier","display_name":"Classifier (UML)","score":0.39079999923706055},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.3736000061035156}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8141999840736389},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6477000117301941},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6409000158309937},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5788999795913696},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5242000222206116},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5084999799728394},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.49900001287460327},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.42399999499320984},{"id":"https://openalex.org/C60908668","wikidata":"https://www.wikidata.org/wiki/Q690207","display_name":"Perceptron","level":3,"score":0.4072999954223633},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.39079999923706055},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.3736000061035156},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.37130001187324524},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.35679998993873596},{"id":"https://openalex.org/C97137487","wikidata":"https://www.wikidata.org/wiki/Q729138","display_name":"Integer (computer science)","level":2,"score":0.35019999742507935},{"id":"https://openalex.org/C519991488","wikidata":"https://www.wikidata.org/wiki/Q28865","display_name":"Python (programming language)","level":2,"score":0.34700000286102295},{"id":"https://openalex.org/C179717631","wikidata":"https://www.wikidata.org/wiki/Q2991667","display_name":"Multilayer perceptron","level":3,"score":0.310699999332428},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.30480000376701355},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.29919999837875366},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.2971999943256378},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.2736999988555908},{"id":"https://openalex.org/C75608658","wikidata":"https://www.wikidata.org/wiki/Q44395","display_name":"Pascal (unit)","level":2,"score":0.273499995470047},{"id":"https://openalex.org/C56086750","wikidata":"https://www.wikidata.org/wiki/Q6042592","display_name":"Integer programming","level":2,"score":0.2680000066757202},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.2628999948501587},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2502000033855438}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.3233/faia250855","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia250855","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},{"id":"pmh:oai:re.public.polimi.it:11311/1308034","is_oa":false,"landing_page_url":"https://hdl.handle.net/11311/1308034","pdf_url":null,"source":{"id":"https://openalex.org/S4306400312","display_name":"Virtual Community of Pathological Anatomy (University of Castilla La Mancha)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79189158","host_organization_name":"University of Castilla-La Mancha","host_organization_lineage":["https://openalex.org/I79189158"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"doi:10.3233/faia250855","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia250855","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Quantization":[0],"is":[1,73,177],"a":[2,96,127,134,157],"pivotal":[3],"technique":[4],"for":[5,87,99,114,173],"managing":[6],"the":[7,20,111,144,206,229],"growing":[8],"computational":[9],"and":[10,28,46,85,117,148,166,217,224],"memory":[11,42,215],"demands":[12],"of":[13,22,49,122],"Deep":[14],"Neural":[15,104],"Networks":[16,105],"(DNNs).":[17],"By":[18],"reducing":[19,214],"number":[21],"bits":[23],"used":[24],"to":[25,35,203,222,228],"represent":[26],"weights":[27],"activations":[29],"(typically":[30],"from":[31],"32-bit":[32],"Floating-Point":[33],"(FP)":[34],"16-bit":[36],"or":[37],"8-bit":[38],"integers),":[39],"quantization":[40,129],"reduces":[41],"footprint,":[43],"energy":[44,218],"consumption,":[45],"execution":[47],"time":[48],"DNNs.":[50],"However,":[51],"most":[52],"existing":[53],"methods":[54],"typically":[55],"target":[56],"DNN":[57],"inference,":[58],"while":[59],"training":[60,100,116,121],"still":[61],"relies":[62],"on":[63,186],"FP":[64,71,231],"operations,":[65],"limiting":[66],"applicability":[67],"in":[68,110],"environments":[69],"where":[70],"arithmetic":[72],"unavailable.":[74],"To":[75],"date,":[76],"only":[77,86],"one":[78],"prior":[79],"work":[80],"has":[81],"addressed":[82],"integer-only":[83,102,195,212],"training,":[84],"Multi-Layer":[88],"Perceptron":[89],"(MLP)":[90],"architectures.":[91],"This":[92],"paper":[93],"introduces":[94,133],"NITRO-D,":[95],"novel":[97,135,158],"framework":[98,154],"deep":[101],"Convolutional":[103],"(CNNs)":[106],"that":[107,137,161],"operate":[108],"entirely":[109],"integer":[112,123,174],"domain":[113],"both":[115],"inference.":[118],"NITRO-D":[119,176,197],"enables":[120],"CNNs":[124],"without":[125],"requiring":[126],"separate":[128],"scheme.":[130],"Specifically,":[131],"it":[132],"architecture":[136],"integrates":[138],"multiple":[139],"local-loss":[140],"blocks,":[141],"which":[142],"include":[143],"proposed":[145,153],"NITRO-Scaling":[146],"layer":[147],"NITRO-ReLU":[149],"activation":[150],"function.":[151],"The":[152],"also":[155,209],"features":[156],"learning":[159],"algorithm":[160],"employs":[162],"local":[163],"error":[164],"signals":[165],"leverages":[167],"IntegerSGD,":[168],"an":[169,180],"optimizer":[170],"specifically":[171],"designed":[172],"computations.":[175],"implemented":[178],"as":[179],"open-source":[181],"Python":[182],"library.":[183],"Extensive":[184],"evaluations":[185],"state-of-the-art":[187],"image":[188],"recognition":[189],"datasets":[190],"demonstrate":[191],"its":[192],"effectiveness.":[193],"For":[194],"MLPs,":[196],"improves":[198],"test":[199],"accuracy":[200],"by":[201,220],"up":[202,221],"+5.96%":[204],"over":[205],"state-of-the-art.":[207],"It":[208],"successfully":[210],"trains":[211],"CNNs,":[213],"requirements":[216],"consumption":[219],"76.14%":[223],"32.42%,":[225],"respectively,":[226],"compared":[227],"traditional":[230],"backpropagation":[232],"algorithm.":[233]},"counts_by_year":[],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-24T00:00:00"}