{"id":"https://openalex.org/W4416251011","doi":"https://doi.org/10.1109/ijcnn64981.2025.11227410","title":"ICE-Pruning: An Iterative Cost-Efficient Pruning Pipeline for Deep Neural Networks","display_name":"ICE-Pruning: An Iterative Cost-Efficient Pruning Pipeline for Deep Neural Networks","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W4416251011","doi":"https://doi.org/10.1109/ijcnn64981.2025.11227410"},"language":"en","primary_location":{"id":"doi:10.1109/ijcnn64981.2025.11227410","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11227410","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://eprints.gla.ac.uk/354072/2/354072.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5084371918","display_name":"Wenhao Hu","orcid":"https://orcid.org/0000-0001-5141-9731"},"institutions":[{"id":"https://openalex.org/I7882870","display_name":"University of Glasgow","ror":"https://ror.org/00vtgdb53","country_code":"GB","type":"education","lineage":["https://openalex.org/I7882870"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Wenhao Hu","raw_affiliation_strings":["University of Glasgow,School of Computing Science,Scotland, UK"],"affiliations":[{"raw_affiliation_string":"University of Glasgow,School of Computing Science,Scotland, UK","institution_ids":["https://openalex.org/I7882870"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013288943","display_name":"Paul Henderson","orcid":"https://orcid.org/0000-0002-5198-7445"},"institutions":[{"id":"https://openalex.org/I7882870","display_name":"University of Glasgow","ror":"https://ror.org/00vtgdb53","country_code":"GB","type":"education","lineage":["https://openalex.org/I7882870"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Paul Henderson","raw_affiliation_strings":["University of Glasgow,School of Computing Science,Scotland, UK"],"affiliations":[{"raw_affiliation_string":"University of Glasgow,School of Computing Science,Scotland, UK","institution_ids":["https://openalex.org/I7882870"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008501980","display_name":"Jos\u00e9 Cano","orcid":"https://orcid.org/0000-0002-2243-389X"},"institutions":[{"id":"https://openalex.org/I7882870","display_name":"University of Glasgow","ror":"https://ror.org/00vtgdb53","country_code":"GB","type":"education","lineage":["https://openalex.org/I7882870"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Jos\u00e9 Cano","raw_affiliation_strings":["University of Glasgow,School of Computing Science,Scotland, UK"],"affiliations":[{"raw_affiliation_string":"University of Glasgow,School of Computing Science,Scotland, UK","institution_ids":["https://openalex.org/I7882870"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5084371918"],"corresponding_institution_ids":["https://openalex.org/I7882870"],"apc_list":null,"apc_paid":null,"fwci":1.242,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.8539608,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.8565000295639038,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.8565000295639038,"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":0.021199999377131462,"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":0.0210999995470047,"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/pruning","display_name":"Pruning","score":0.9326000213623047},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.6377999782562256},{"id":"https://openalex.org/keywords/hyperparameter","display_name":"Hyperparameter","score":0.6312999725341797},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6157000064849854},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4068000018596649},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.38850000500679016}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.9326000213623047},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7372999787330627},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6377999782562256},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.6312999725341797},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6157000064849854},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5121999979019165},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.43459999561309814},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4068000018596649},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.38850000500679016},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.364300012588501},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3138999938964844},{"id":"https://openalex.org/C114290370","wikidata":"https://www.wikidata.org/wiki/Q1415915","display_name":"Null-move heuristic","level":3,"score":0.29179999232292175},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.28380000591278076},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.26249998807907104},{"id":"https://openalex.org/C175309249","wikidata":"https://www.wikidata.org/wiki/Q725864","display_name":"Pipeline transport","level":2,"score":0.25429999828338623}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/ijcnn64981.2025.11227410","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11227410","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},{"id":"pmh:oai:eprints.gla.ac.uk:354072","is_oa":true,"landing_page_url":"https://eprints.gla.ac.uk/view/author/62797.html>,","pdf_url":"https://eprints.gla.ac.uk/354072/2/354072.pdf","source":{"id":"https://openalex.org/S4210235606","display_name":"ENLIGHTEN (Jurnal Bimbingan dan Konseling Islam)","issn_l":"2622-8912","issn":["2622-8912","2622-8920"],"is_oa":true,"is_in_doaj":true,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"PeerReviewed"},{"id":"pmh:oai:eprints.gla.ac.uk:354072","is_oa":true,"landing_page_url":"http://eprints.gla.ac.uk/view/author/62797.html>,","pdf_url":null,"source":{"id":"https://openalex.org/S4210235606","display_name":"ENLIGHTEN (Jurnal Bimbingan dan Konseling Islam)","issn_l":"2622-8912","issn":["2622-8912","2622-8920"],"is_oa":true,"is_in_doaj":true,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"PeerReviewed"}],"best_oa_location":{"id":"pmh:oai:eprints.gla.ac.uk:354072","is_oa":true,"landing_page_url":"https://eprints.gla.ac.uk/view/author/62797.html>,","pdf_url":"https://eprints.gla.ac.uk/354072/2/354072.pdf","source":{"id":"https://openalex.org/S4210235606","display_name":"ENLIGHTEN (Jurnal Bimbingan dan Konseling Islam)","issn_l":"2622-8912","issn":["2622-8912","2622-8920"],"is_oa":true,"is_in_doaj":true,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"PeerReviewed"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416251011.pdf","grobid_xml":"https://content.openalex.org/works/W4416251011.grobid-xml"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W2108598243","https://openalex.org/W2194775991","https://openalex.org/W2808168148","https://openalex.org/W2949676527","https://openalex.org/W2963446712","https://openalex.org/W2964137095","https://openalex.org/W2964233199","https://openalex.org/W3012561096","https://openalex.org/W3041571781","https://openalex.org/W3091661482","https://openalex.org/W3174700868","https://openalex.org/W4365420873","https://openalex.org/W4402134764","https://openalex.org/W4407893752"],"related_works":[],"abstract_inverted_index":{"Pruning":[0],"is":[1,34,53,127,219],"a":[2,19,119,148,160],"widely":[3],"used":[4],"method":[5],"for":[6,88,100,108,151,188],"compressing":[7],"Deep":[8],"Neural":[9],"Networks":[10],"(DNNs),":[11],"where":[12],"less":[13],"relevant":[14],"parameters":[15,28,65],"are":[16,81],"removed":[17,67],"from":[18],"DNN":[20,204],"model":[21,30,64],"to":[22,47,54,73,85,122,137,166,216],"reduce":[23,176],"its":[24],"size.":[25],"However,":[26],"removing":[27],"reduces":[29],"accuracy,":[31],"so":[32],"pruning":[33,75,98,109,124,141,156,173,213],"typically":[35],"combined":[36],"with":[37,60],"fine-tuning,":[38,116],"and":[39,57,158,175,206],"sometimes":[40],"other":[41],"operations":[42],"such":[43],"as":[44],"rewinding":[45],"weights,":[46],"recover":[48],"accuracy.":[49],"A":[50],"common":[51],"approach":[52,80],"repeatedly":[55],"prune":[56],"then":[58],"fine-tune,":[59],"increasing":[61],"amounts":[62],"of":[63,115,171],"being":[66],"in":[68,154],"each":[69,155,172],"step.":[70],"While":[71],"straightforward":[72],"implement,":[74],"pipelines":[76],"that":[77,102,209],"follow":[78],"this":[79,91],"computationally":[82],"expensive":[83],"due":[84],"the":[86,105,112,169,177,189,196],"need":[87],"repeated":[89],"fine-tuning.In":[90],"paper":[92],"we":[93],"propose":[94,183],"ICE-Pruning,":[95],"an":[96,134,184],"iterative":[97],"pipeline":[99],"DNNs":[101],"significantly":[103],"decreases":[104],"time":[106,179],"required":[107],"by":[110,195,214],"reducing":[111],"overall":[113,178],"cost":[114],"while":[117],"maintaining":[118],"similar":[120],"accuracy":[121,170],"existing":[123],"pipelines.":[125],"ICE-Pruning":[126,201],"based":[128],"on":[129,202],"three":[130,197],"main":[131],"components:":[132],"i)":[133],"automatic":[135],"mechanism":[136],"determine":[138],"after":[139],"which":[140],"steps":[142],"fine-tuning":[143,153],"should":[144],"be":[145],"performed;":[146],"ii)":[147],"freezing":[149,192],"strategy":[150],"faster":[152],"step;":[157],"iii)":[159],"custom":[161],"pruning-aware":[162],"learning":[163],"rate":[164],"scheduler":[165],"further":[167],"improve":[168],"step":[174],"consumption.":[180],"We":[181,199],"also":[182],"efficient":[185],"auto-tuning":[186],"stage":[187],"hyperparameters":[190],"(e.g.,":[191],"percentage)":[193],"introduced":[194],"components.":[198],"evaluate":[200],"several":[203],"models":[205],"datasets,":[207],"showing":[208],"it":[210],"can":[211],"accelerate":[212],"up":[215],"9.61\u00d7.":[217],"Code":[218],"available":[220],"at":[221],"https://github.com/gicLAB/ICE-Pruning":[222]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}