{"id":"https://openalex.org/W2897110360","doi":null,"title":"Pruning neural networks: is it time to nip it in the bud?","display_name":"Pruning neural networks: is it time to nip it in the bud?","publication_year":2018,"publication_date":"2018-10-10","ids":{"openalex":"https://openalex.org/W2897110360","mag":"2897110360"},"language":"en","primary_location":{"id":"mag:2897110360","doi":null,"is_oa":true,"landing_page_url":"https://openreview.net/pdf?id=r1lbgwFj5m","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://openreview.net/pdf?id=r1lbgwFj5m","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5003510948","display_name":"Elliot J. Crowley","orcid":"https://orcid.org/0000-0001-5685-4724","observed_orcids":["https://orcid.org/0000-0001-5685-4724"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Elliot J. Crowley","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0001-5685-4724","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049917315","display_name":"Jack Turner","orcid":"https://orcid.org/0000-0003-4470-1805","observed_orcids":["https://orcid.org/0000-0003-4470-1805"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jack Turner","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0003-4470-1805","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007901825","display_name":"Amos Storkey","orcid":"https://orcid.org/0000-0002-8100-506X","observed_orcids":["https://orcid.org/0000-0002-8100-506X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Amos Storkey","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5027001025","display_name":"Michael F. P. O\u2019boyle","orcid":"https://orcid.org/0000-0003-1619-5052","observed_orcids":["https://orcid.org/0000-0003-1619-5052"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Michael O\u2019Boyle","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0003-1619-5052","affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":20,"citation_normalized_percentile":null,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9994999766349792,"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":0.9994999766349792,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.00019999999494757503,"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":9.999999747378752e-05,"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/neural-network-pruning","display_name":"neural network pruning","score":0.9940000176429749},{"id":"https://openalex.org/keywords/model-compression","display_name":"model compression","score":0.6940000057220459},{"id":"https://openalex.org/keywords/resnet","display_name":"ResNet","score":0.7730000019073486},{"id":"https://openalex.org/keywords/densenet","display_name":"DenseNet","score":0.9990000128746033},{"id":"https://openalex.org/keywords/training-from-scratch","display_name":"training from scratch","score":0.8730000257492065},{"id":"https://openalex.org/keywords/inference-speed","display_name":"inference speed","score":0.9049999713897705},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"hardware acceleration","score":0.3709999918937683}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.901023268699646},{"id":"https://openalex.org/C2781235140","wikidata":"https://www.wikidata.org/wiki/Q275131","display_name":"Scratch","level":2,"score":0.8544352054595947},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.779563307762146},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6834491491317749},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6541811227798462},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5561262965202332},{"id":"https://openalex.org/C193415008","wikidata":"https://www.wikidata.org/wiki/Q639681","display_name":"Network architecture","level":2,"score":0.4901674687862396},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.40483370423316956},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.2335016429424286},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:2897110360","doi":null,"is_oa":true,"landing_page_url":"https://openreview.net/pdf?id=r1lbgwFj5m","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null}],"best_oa_location":{"id":"mag:2897110360","doi":null,"is_oa":true,"landing_page_url":"https://openreview.net/pdf?id=r1lbgwFj5m","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W1821462560","https://openalex.org/W2114766824","https://openalex.org/W2117539524","https://openalex.org/W2120972216","https://openalex.org/W2123045220","https://openalex.org/W2134797427","https://openalex.org/W2194775991","https://openalex.org/W2286929393","https://openalex.org/W2331143823","https://openalex.org/W2460144244","https://openalex.org/W2513419314","https://openalex.org/W2541674938","https://openalex.org/W2561238782","https://openalex.org/W2604319603","https://openalex.org/W2619890685","https://openalex.org/W2736941579","https://openalex.org/W2765720631","https://openalex.org/W2768607686","https://openalex.org/W2783873922","https://openalex.org/W2786054724","https://openalex.org/W2810075754","https://openalex.org/W2886851211","https://openalex.org/W2890392327","https://openalex.org/W2894740066","https://openalex.org/W2895171208","https://openalex.org/W2896409484","https://openalex.org/W2915589364","https://openalex.org/W2919115771","https://openalex.org/W2947776571","https://openalex.org/W2949117887","https://openalex.org/W2953212265","https://openalex.org/W2962819303","https://openalex.org/W2962861284","https://openalex.org/W2962935923","https://openalex.org/W2963363373","https://openalex.org/W2963446712","https://openalex.org/W2963655672","https://openalex.org/W2963674932","https://openalex.org/W2963813662","https://openalex.org/W2964137095","https://openalex.org/W2964299589","https://openalex.org/W2991391304","https://openalex.org/W3118608800"],"related_works":["https://openalex.org/W2948879487","https://openalex.org/W3005165565","https://openalex.org/W3163244872","https://openalex.org/W3209135316","https://openalex.org/W4281650003","https://openalex.org/W3092116469","https://openalex.org/W3000076995","https://openalex.org/W7070987804","https://openalex.org/W6976457947","https://openalex.org/W3126027909"],"abstract_inverted_index":{"Pruning":[0],"is":[1,86],"a":[2,7,11,76,103],"popular":[3],"method":[4],"for":[5],"compressing":[6],"neural":[8],"network:":[9],"given":[10],"large":[12],"trained":[13,63,115],"network,":[14],"one":[15],"alternates":[16],"between":[17],"removing":[18],"connections":[19],"and":[20,45,50,79,101,126,131],"fine-tuning;":[21],"reducing":[22],"the":[23,27,30,60,73,121],"overall":[24],"width":[25],"of":[26,32,59,75,106,124],"network.":[28],"However,":[29],"efficacy":[31],"network":[33,62,78,109],"pruning":[34],"has":[35],"largely":[36],"evaded":[37],"scrutiny.":[38],"In":[39],"this":[40],"paper,":[41],"we":[42,97,119],"examine":[43],"ResNets":[44],"DenseNets":[46],"obtained":[47],"through":[48],"pruning-and-tuning":[49],"make":[51],"two":[52],"interesting":[53],"observations:":[54],"(i)":[55],"reduced":[56,125,134],"networks---smaller":[57],"versions":[58],"original":[61],"from":[64,83,116],"scratch---consistently":[65],"outperform":[66],"pruned":[67,77,127],"networks;":[68],"(ii)":[69],"if":[70],"you":[71],"take":[72],"architecture":[74],"then":[80],"train":[81],"it":[82,85],"scratch":[84],"significantly":[87,137],"more":[88],"competitive.":[89],"Furthermore,":[90],"these":[91],"architectures":[92,110],"are":[93,136],"easy":[94],"to":[95],"approximate:":[96],"can":[98,112],"prune":[99],"once":[100],"obtain":[102],"whole":[104],"family":[105],"new,":[107],"scalable":[108],"that":[111,133],"simply":[113],"be":[114],"scratch.":[117],"Finally,":[118],"compare":[120],"inference":[122],"speed":[123],"networks":[128,135],"on":[129],"hardware,":[130],"show":[132],"faster.":[138]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":6},{"year":2020,"cited_by_count":6},{"year":2019,"cited_by_count":5},{"year":2018,"cited_by_count":1}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2018-10-26T00:00:00"}