{"id":"https://openalex.org/W7106125252","doi":"https://doi.org/10.1109/iccad66269.2025.11240988","title":"Perturbation-efficient Zeroth-order Optimization for Hardware-friendly On-device Training","display_name":"Perturbation-efficient Zeroth-order Optimization for Hardware-friendly On-device Training","publication_year":2025,"publication_date":"2025-10-26","ids":{"openalex":"https://openalex.org/W7106125252","doi":"https://doi.org/10.1109/iccad66269.2025.11240988"},"language":null,"primary_location":{"id":"doi:10.1109/iccad66269.2025.11240988","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad66269.2025.11240988","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","raw_type":"proceedings-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":null,"display_name":"Qitao Tan","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Qitao Tan","raw_affiliation_strings":["University of Georgia"],"affiliations":[{"raw_affiliation_string":"University of Georgia","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Sung-En Chang","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Sung-En Chang","raw_affiliation_strings":["Northeastern University"],"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Rui Xia","orcid":null},"institutions":[{"id":"https://openalex.org/I36788626","display_name":"California University of Pennsylvania","ror":"https://ror.org/01spssf70","country_code":"US","type":"education","lineage":["https://openalex.org/I36788626"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rui Xia","raw_affiliation_strings":["University of Pennsylvania"],"affiliations":[{"raw_affiliation_string":"University of Pennsylvania","institution_ids":["https://openalex.org/I36788626"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Huidong Ji","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huidong Ji","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Chence Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chence Yang","raw_affiliation_strings":["University of Georgia"],"affiliations":[{"raw_affiliation_string":"University of Georgia","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Ci Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ci Zhang","raw_affiliation_strings":["University of Georgia"],"affiliations":[{"raw_affiliation_string":"University of Georgia","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jun Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Jun Liu","raw_affiliation_strings":["Northeastern University"],"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zheng Zhan","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Zheng Zhan","raw_affiliation_strings":["Northeastern University"],"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zhenman Fang","orcid":null},"institutions":[{"id":"https://openalex.org/I18014758","display_name":"Simon Fraser University","ror":"https://ror.org/0213rcc28","country_code":"CA","type":"education","lineage":["https://openalex.org/I18014758"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Zhenman Fang","raw_affiliation_strings":["Simon Fraser University"],"affiliations":[{"raw_affiliation_string":"Simon Fraser University","institution_ids":["https://openalex.org/I18014758"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zhuo Zou","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhuo Zou","raw_affiliation_strings":["Fudan University"],"affiliations":[{"raw_affiliation_string":"Fudan University","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yanzhi Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Yanzhi Wang","raw_affiliation_strings":["Northeastern University"],"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jin Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jin Lu","raw_affiliation_strings":["University of Georgia"],"affiliations":[{"raw_affiliation_string":"University of Georgia","institution_ids":[]}]},{"author_position":"last","author":{"id":null,"display_name":"Geng Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Geng Yuan","raw_affiliation_strings":["University of Georgia"],"affiliations":[{"raw_affiliation_string":"University of Georgia","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":13,"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":0.57897811,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11017","display_name":"Chaos-based Image/Signal Encryption","score":0.15240000188350677,"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"}},"topics":[{"id":"https://openalex.org/T11017","display_name":"Chaos-based Image/Signal Encryption","score":0.15240000188350677,"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/T11697","display_name":"Numerical Methods and Algorithms","score":0.14910000562667847,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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":0.0502999983727932,"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/reuse","display_name":"Reuse","score":0.5497999787330627},{"id":"https://openalex.org/keywords/random-number-generation","display_name":"Random number generation","score":0.536899983882904},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.48249998688697815},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.4618000090122223},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.45509999990463257},{"id":"https://openalex.org/keywords/optimization-problem","display_name":"Optimization problem","score":0.4185999929904938},{"id":"https://openalex.org/keywords/power","display_name":"Power (physics)","score":0.3626999855041504},{"id":"https://openalex.org/keywords/simplicity","display_name":"Simplicity","score":0.3483999967575073}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7253999710083008},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.5497999787330627},{"id":"https://openalex.org/C201866948","wikidata":"https://www.wikidata.org/wiki/Q228206","display_name":"Random number generation","level":2,"score":0.536899983882904},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.48249998688697815},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.4618000090122223},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.45509999990463257},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4237000048160553},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.4185999929904938},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4007999897003174},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39579999446868896},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3756999969482422},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.3626999855041504},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35019999742507935},{"id":"https://openalex.org/C2776372474","wikidata":"https://www.wikidata.org/wiki/Q508291","display_name":"Simplicity","level":2,"score":0.3483999967575073},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.34439998865127563},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.3325999975204468},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.3174999952316284},{"id":"https://openalex.org/C126661757","wikidata":"https://www.wikidata.org/wiki/Q4925641","display_name":"Random search","level":2,"score":0.3046000003814697},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.29840001463890076},{"id":"https://openalex.org/C2987595161","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Optimization algorithm","level":2,"score":0.2906999886035919},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C168292644","wikidata":"https://www.wikidata.org/wiki/Q10860336","display_name":"Power optimization","level":4,"score":0.2621000111103058},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25870001316070557}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iccad66269.2025.11240988","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad66269.2025.11240988","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","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":23,"referenced_works":["https://openalex.org/W2028175314","https://openalex.org/W2060615856","https://openalex.org/W2097774686","https://openalex.org/W2130158090","https://openalex.org/W2142840816","https://openalex.org/W2251939518","https://openalex.org/W2736068458","https://openalex.org/W2912949188","https://openalex.org/W2950598802","https://openalex.org/W3092347650","https://openalex.org/W3114218318","https://openalex.org/W3138582970","https://openalex.org/W3173777717","https://openalex.org/W3177323791","https://openalex.org/W3179431530","https://openalex.org/W4241542937","https://openalex.org/W4244885949","https://openalex.org/W4253067820","https://openalex.org/W4312250727","https://openalex.org/W4393578753","https://openalex.org/W4401568214","https://openalex.org/W4402742487","https://openalex.org/W4404782338"],"related_works":[],"abstract_inverted_index":{"Zeroth-order":[0],"(ZO)":[1],"optimization":[2,152],"is":[3],"an":[4],"emerging":[5],"deep":[6],"neural":[7],"network":[8],"(DNN)":[9],"training":[10,148],"paradigm":[11],"that":[12,122],"offers":[13],"computational":[14],"simplicity":[15],"and":[16,27,44,55,72,102,128,136,138],"memory":[17],"savings.":[18],"However,":[19],"this":[20,58,62,77],"seemingly":[21],"promising":[22],"approach":[23],"faces":[24],"a":[25,33,82,104,116],"significant":[26,42],"long-ignored":[28],"challenge.":[29],"ZO":[30,84,151,173],"requires":[31],"generating":[32],"substantial":[34],"number":[35,90,100,132],"of":[36,160,171],"Gaussian":[37,113],"random":[38,89,99,131],"numbers,":[39],"which":[40,65],"poses":[41],"difficulties":[43],"even":[45],"makes":[46],"it":[47],"infeasible":[48],"for":[49,98,130,154,178],"hardware":[50,73],"platforms,":[51],"such":[52],"as":[53],"FPGAs":[54],"ASICs.":[56],"In":[57],"paper,":[59],"we":[60,79,87,163],"identify":[61],"critical":[63],"issue,":[64,78],"arises":[66],"from":[67],"the":[68,96,111,125,158,165,169],"mismatch":[69],"between":[70],"algorithm":[71],"designers.":[74],"To":[75,157],"address":[76],"proposed":[80],"PeZO,":[81],"perturbation-efficient":[83],"framework.":[85],"Specifically,":[86],"design":[88],"reuse":[91],"strategies":[92],"to":[93,109,167],"significantly":[94],"reduce":[95],"demand":[97],"generation":[101,133],"introduce":[103],"hardware-friendly":[105],"adaptive":[106],"scaling":[107],"method":[108],"replace":[110],"costly":[112],"distribution":[114],"with":[115],"uniform":[117],"distribution.":[118],"Our":[119],"experiments":[120],"show":[121],"PeZO":[123],"reduces":[124],"required":[126],"LUTs":[127],"FFs":[129],"by":[134],"48.6%":[135],"12.7%,":[137],"saves":[139],"at":[140],"maximum":[141],"86%":[142],"power":[143],"consumption,":[144],"all":[145],"without":[146],"compromising":[147],"performance,":[149],"making":[150],"feasible":[153],"on-device":[155,172],"training.":[156],"best":[159],"our":[161],"knowledge,":[162],"are":[164],"first":[166],"explore":[168],"potential":[170],"optimization,":[174],"providing":[175],"valuable":[176],"insights":[177],"future":[179],"research.":[180]},"counts_by_year":[],"updated_date":"2025-11-23T05:10:03.516525","created_date":"2025-11-20T00:00:00"}