{"id":"https://openalex.org/W7162665582","doi":"https://doi.org/10.1109/access.2026.3697896","title":"Path-Averaged Gradient Approximation Algorithms for Deep Neural Network Training","display_name":"Path-Averaged Gradient Approximation Algorithms for Deep Neural Network Training","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7162665582","doi":"https://doi.org/10.1109/access.2026.3697896"},"language":"en","primary_location":{"id":"doi:10.1109/access.2026.3697896","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3697896","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1109/access.2026.3697896","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5092719977","display_name":"Rafa\u0142 Wolniak","orcid":null},"institutions":[{"id":"https://openalex.org/I169333911","display_name":"Gda\u0144sk University of Technology","ror":"https://ror.org/006x4sc24","country_code":"PL","type":"education","lineage":["https://openalex.org/I169333911"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Rafal Wolniak","raw_affiliation_strings":["Multimedia Systems Department, Audio Acoustics Laboratory, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, Gdańsk, Poland"],"raw_orcid":"https://orcid.org/0009-0004-8144-5315","affiliations":[{"raw_affiliation_string":"Multimedia Systems Department, Audio Acoustics Laboratory, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, Gdańsk, Poland","institution_ids":["https://openalex.org/I169333911"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024105413","display_name":"Adam Kurowski","orcid":"https://orcid.org/0000-0002-5132-3016"},"institutions":[{"id":"https://openalex.org/I169333911","display_name":"Gda\u0144sk University of Technology","ror":"https://ror.org/006x4sc24","country_code":"PL","type":"education","lineage":["https://openalex.org/I169333911"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Adam Kurowski","raw_affiliation_strings":["Multimedia Systems Department, Audio Acoustics Laboratory, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, Gdańsk, Poland"],"raw_orcid":"https://orcid.org/0000-0002-5132-3016","affiliations":[{"raw_affiliation_string":"Multimedia Systems Department, Audio Acoustics Laboratory, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, Gdańsk, Poland","institution_ids":["https://openalex.org/I169333911"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033154448","display_name":"Bo\u017cena Kostek","orcid":"https://orcid.org/0000-0001-6288-2908"},"institutions":[{"id":"https://openalex.org/I169333911","display_name":"Gda\u0144sk University of Technology","ror":"https://ror.org/006x4sc24","country_code":"PL","type":"education","lineage":["https://openalex.org/I169333911"]}],"countries":["PL"],"is_corresponding":false,"raw_author_name":"Bozena Kostek","raw_affiliation_strings":["Multimedia Systems Department, Audio Acoustics Laboratory, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, Gdańsk, Poland"],"raw_orcid":"https://orcid.org/0000-0001-6288-2908","affiliations":[{"raw_affiliation_string":"Multimedia Systems Department, Audio Acoustics Laboratory, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, Gdańsk, Poland","institution_ids":["https://openalex.org/I169333911"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I169333911"],"apc_list":{"value":1850,"currency":"USD","value_usd":1850},"apc_paid":{"value":1850,"currency":"USD","value_usd":1850},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.66331195,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"14","issue":null,"first_page":"88765","last_page":"88785"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.447299987077713,"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":0.447299987077713,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.06960000097751617,"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/T12676","display_name":"Machine Learning and ELM","score":0.06809999793767929,"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/artificial-neural-network","display_name":"Artificial neural network","score":0.6083999872207642},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.4918999969959259},{"id":"https://openalex.org/keywords/approximation-algorithm","display_name":"Approximation algorithm","score":0.4481000006198883},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.43630000948905945},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.36329999566078186},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3490999937057495},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3230000138282776}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7373999953269958},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6083999872207642},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5457000136375427},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5407999753952026},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.4918999969959259},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.4481000006198883},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.43630000948905945},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.36329999566078186},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3490999937057495},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3230000138282776},{"id":"https://openalex.org/C145242015","wikidata":"https://www.wikidata.org/wiki/Q774123","display_name":"Approximation theory","level":2,"score":0.31529998779296875},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.2953999936580658},{"id":"https://openalex.org/C106516650","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm design","level":2,"score":0.2944999933242798},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.29339998960494995},{"id":"https://openalex.org/C175202392","wikidata":"https://www.wikidata.org/wiki/Q2434543","display_name":"Time delay neural network","level":3,"score":0.2759999930858612},{"id":"https://openalex.org/C47702885","wikidata":"https://www.wikidata.org/wiki/Q5441227","display_name":"Feedforward neural network","level":3,"score":0.27410000562667847},{"id":"https://openalex.org/C115680565","wikidata":"https://www.wikidata.org/wiki/Q5977448","display_name":"Gradient method","level":2,"score":0.26109999418258667},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.25760000944137573}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/access.2026.3697896","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3697896","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:470a36c29fa249d892eb8fb513fb1bda","is_oa":false,"landing_page_url":"https://doaj.org/article/470a36c29fa249d892eb8fb513fb1bda","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Access, Vol 14, Pp 88765-88785 (2026)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1109/access.2026.3697896","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3697896","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"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":{"We":[0,46,69,99,132],"explore":[1],"previously":[2],"unreported":[3],"properties":[4],"and":[5,92,141,157,160],"practical":[6],"utility":[7],"of":[8,35,104,116],"integrating":[9],"gradients":[10,30,113,150],"along":[11],"parameter":[12,36],"trajectories":[13],"for":[14,72,108,155],"training":[15,118],"deep":[16,164],"neural":[17,165],"networks.":[18,166],"Unlike":[19],"standard":[20,84],"methods":[21],"that":[22,71,111],"rely":[23],"on":[24,120],"instantaneous":[25],"gradients,":[26],"this":[27,53],"approach":[28],"averages":[29],"over":[31,114],"a":[32,73,76,121,152],"continuous":[33],"range":[34],"values":[37],"at":[38,124],"each":[39,125],"update":[40],"step.":[41],"Our":[42],"contributions":[43,135],"are:":[44],"(a)":[45],"show":[47],"that,":[48],"across":[49,130],"multiple":[50],"convolutional":[51],"architectures,":[52],"method":[54],"yields":[55],"up":[56],"to":[57,65,95],"53.5%":[58],"greater":[59],"reduction":[60],"in":[61,87,163],"per-batch":[62],"loss":[63],"compared":[64],"baseline":[66],"optimizers.":[67,131,144],"(b)":[68],"demonstrate":[70],"fixed":[74,122],"batch,":[75],"single":[77],"step":[78],"can":[79],"approximate":[80],"more":[81],"than":[82],"four":[83],"gradient":[85,107],"steps":[86],"models":[88],"lacking":[89],"skip":[90],"connections":[91],"thus":[93],"prone":[94],"ill-conditioned":[96],"curvature.":[97],"(c)":[98],"introduce":[100],"an":[101],"efficient":[102],"approximation":[103],"the":[105],"average":[106],"ResNet-152":[109],"fine-tuning":[110],"aggregates":[112],"hundreds":[115],"past":[117],"iterations":[119],"batch":[123],"update,":[126],"yielding":[127],"improved":[128],"accuracy":[129],"validate":[133],"these":[134],"using":[136],"representative":[137],"first-order":[138],"(RMSProp,":[139],"Adam)":[140],"second-order":[142],"(SOAP)":[143],"These":[145],"results":[146],"highlight":[147],"parameter-space":[148],"integrated":[149],"as":[151],"novel":[153],"framework":[154],"understanding":[156],"improving":[158],"generalization":[159],"local-descent":[161],"efficiency":[162]},"counts_by_year":[],"updated_date":"2026-06-18T08:10:14.011955","created_date":"2026-05-29T00:00:00"}