{"id":"https://openalex.org/W3036842007","doi":"https://doi.org/10.48550/arxiv.2006.11967","title":"Exploiting Weight Redundancy in CNNs: Beyond Pruning and Quantization","display_name":"Exploiting Weight Redundancy in CNNs: Beyond Pruning and Quantization","alternate_titles":[],"publication_year":2020,"publication_date":"2020-06-22","ids":{"openalex":"https://openalex.org/W3036842007","doi":"https://doi.org/10.48550/arxiv.2006.11967","mag":"3036842007"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2006.11967","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2006.11967","pdf_url":"https://arxiv.org/pdf/2006.11967","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":null,"raw_type":"text"},"parent_works":[],"lineage":["https://openalex.org/W3036842007"],"child_works_count":0,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2006.11967","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101113814","display_name":"Yuan Wen","orcid":"https://orcid.org/0000-0001-5973-1519","observed_orcids":["https://orcid.org/0000-0001-5973-1519"]},"institutions":[{"id":"https://openalex.org/I205274468","display_name":"Trinity College Dublin","ror":"https://ror.org/02tyrky19","country_code":"IE","type":"education","lineage":["https://openalex.org/I205274468"]}],"countries":["IE"],"is_corresponding":false,"contributor_roles":[],"raw_author_name":"Wen, Yuan","raw_affiliation_strings":["Trinity College Dublin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Trinity College Dublin","institution_ids":["https://openalex.org/I205274468"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5003800161","display_name":"David Gregg","orcid":"https://orcid.org/0000-0003-3782-4612","observed_orcids":["https://orcid.org/0000-0003-3782-4612"]},"institutions":[{"id":"https://openalex.org/I205274468","display_name":"Trinity College Dublin","ror":"https://ror.org/02tyrky19","country_code":"IE","type":"education","lineage":["https://openalex.org/I205274468"]}],"countries":["IE"],"is_corresponding":false,"contributor_roles":[],"raw_author_name":"Gregg, David","raw_affiliation_strings":["Trinity College Dublin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Trinity College Dublin","institution_ids":["https://openalex.org/I205274468"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I205274468"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":9,"reviewed_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_pulled":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9961000084877014,"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":0.9961000084877014,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.0019000000320374966,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.00019999999494757503,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/weight-redundancy","display_name":"weight redundancy","score":0.7680000066757202},{"id":"https://openalex.org/keywords/convolutional-neural-networks","display_name":"convolutional neural networks","score":0.9229999780654907},{"id":"https://openalex.org/keywords/neural-network-compression","display_name":"neural network compression","score":0.8309999704360962},{"id":"https://openalex.org/keywords/huffman-coding","display_name":"Huffman coding","score":0.9100000262260437},{"id":"https://openalex.org/keywords/weight-pruning","display_name":"weight pruning","score":0.5550000071525574},{"id":"https://openalex.org/keywords/weight-quantization","display_name":"weight quantization","score":0.9240000247955322},{"id":"https://openalex.org/keywords/storage-efficiency","display_name":"storage efficiency","score":0.6800000071525574}],"concepts":[{"id":"https://openalex.org/C46900642","wikidata":"https://www.wikidata.org/wiki/Q2647","display_name":"Huffman coding","level":3,"score":0.7442749738693237},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7099408507347107},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6159400343894958},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5975232720375061},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.584035336971283},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5048211216926575},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.38434717059135437},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3831912875175476},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3123741149902344},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.3029414117336273},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2006.11967","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2006.11967","pdf_url":"https://arxiv.org/pdf/2006.11967","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":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2006.11967","doi":"https://doi.org/10.48550/arxiv.2006.11967","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2006.11967","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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2006.11967","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2006.11967","pdf_url":"https://arxiv.org/pdf/2006.11967","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":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3036842007.pdf","grobid_xml":"https://content.openalex.org/works/W3036842007.grobid-xml"},"referenced_works_count":28,"referenced_works":["https://openalex.org/W107968503","https://openalex.org/W587794757","https://openalex.org/W1506342804","https://openalex.org/W1527464046","https://openalex.org/W1845051632","https://openalex.org/W2106721790","https://openalex.org/W2114766824","https://openalex.org/W2119144962","https://openalex.org/W2125389748","https://openalex.org/W2161591461","https://openalex.org/W2276892413","https://openalex.org/W2337344472","https://openalex.org/W2420699360","https://openalex.org/W2483966489","https://openalex.org/W2495425901","https://openalex.org/W2507318699","https://openalex.org/W2515385951","https://openalex.org/W2554302513","https://openalex.org/W2604319603","https://openalex.org/W2657126969","https://openalex.org/W2736953746","https://openalex.org/W2743799610","https://openalex.org/W2751366252","https://openalex.org/W2752037867","https://openalex.org/W2798544842","https://openalex.org/W2963363373","https://openalex.org/W2964233199","https://openalex.org/W4394670654"],"related_works":["https://openalex.org/W3196485035","https://openalex.org/W2892614785","https://openalex.org/W3042992683","https://openalex.org/W2970805313","https://openalex.org/W2953974830","https://openalex.org/W3164441165","https://openalex.org/W3094557303","https://openalex.org/W2896811383","https://openalex.org/W4288043077"],"abstract_inverted_index":{"Pruning":[0,18],"and":[1,10,24,76,113,130,149],"quantization":[2,77],"are":[3],"proven":[4],"methods":[5],"for":[6],"improving":[7],"the":[8,35,64,83,89,145],"performance":[9],"storage":[11],"efficiency":[12],"of":[13,37,57,66,69,85,100,110,138],"convolutional":[14],"neural":[15],"networks":[16],"(CNNs).":[17],"removes":[19],"near-zero":[20],"weights":[21,38],"in":[22,30,59,63,88,103,142],"tensors":[23],"masks":[25],"weak":[26],"connections":[27],"between":[28],"neurons":[29],"neighbouring":[31],"layers.":[32],"Quantization":[33],"reduces":[34],"precision":[36],"by":[39,117],"replacing":[40],"them":[41],"with":[42],"numerically":[43],"similar":[44,70],"values":[45],"that":[46,74,132],"require":[47],"less":[48],"storage.":[49],"In":[50],"this":[51,101],"paper,":[52],"we":[53,133],"identify":[54],"another":[55],"form":[56,65],"redundancy":[58],"CNN":[60,104],"weight":[61,90,105],"tensors,":[62],"repeated":[67,86],"patterns":[68,87],"values.":[71],"We":[72,92,122],"observe":[73],"pruning":[75,148],"both":[78],"tend":[79],"to":[80,97,140,144],"drastically":[81],"increase":[82],"number":[84],"tensors.":[91],"investigate":[93],"several":[94,127],"compression":[95],"schemes":[96],"take":[98],"advantage":[99],"structure":[102],"data,":[106],"including":[107],"multiple":[108],"forms":[109],"Huffman":[111],"coding,":[112],"other":[114],"approaches":[115],"inspired":[116],"block":[118],"sparse":[119],"matrix":[120],"formats.":[121],"evaluate":[123],"our":[124],"approach":[125],"on":[126],"well-known":[128],"CNNs":[129],"find":[131],"can":[134],"achieve":[135],"compaction":[136],"ratios":[137],"1.4x":[139],"3.1x":[141],"addition":[143],"saving":[146],"from":[147],"quantization.":[150]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W3036842007","counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":2}],"updated_date":"2026-11-18T13:11:19.462570","created_date":"2020-06-25T00:00:00"}
