{"id":"https://openalex.org/W3129112952","doi":null,"title":"GNN-RL Compression: Topology-Aware Network Pruning using Multi-stage Graph Embedding and Reinforcement Learning.","display_name":"GNN-RL Compression: Topology-Aware Network Pruning using Multi-stage Graph Embedding and Reinforcement Learning.","publication_year":2021,"publication_date":"2021-02-05","ids":{"openalex":"https://openalex.org/W3129112952","mag":"3129112952"},"language":"en","primary_location":{"id":"mag:3129112952","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2102.03214.pdf","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://arxiv.org/pdf/2102.03214.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5085171738","display_name":"Sixing Yu","orcid":"https://orcid.org/0000-0002-2415-566X","observed_orcids":["https://orcid.org/0000-0002-2415-566X"]},"institutions":[{"id":"https://openalex.org/I173911158","display_name":"Iowa State University","ror":"https://ror.org/04rswrd78","country_code":"US","type":"education","lineage":["https://openalex.org/I173911158"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sixing Yu","raw_affiliation_strings":["Iowa State University\u2028"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Iowa State University\u2028","institution_ids":["https://openalex.org/I173911158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028887858","display_name":"Arya Mazaheri","orcid":"https://orcid.org/0000-0002-5671-4710","observed_orcids":["https://orcid.org/0000-0002-5671-4710"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Arya Mazaheri","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-5671-4710","affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5079359777","display_name":"Ali Jannesari","orcid":"https://orcid.org/0000-0001-8672-5317","observed_orcids":["https://orcid.org/0000-0001-8672-5317"]},"institutions":[{"id":"https://openalex.org/I173911158","display_name":"Iowa State University","ror":"https://ror.org/04rswrd78","country_code":"US","type":"education","lineage":["https://openalex.org/I173911158"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ali Jannesari","raw_affiliation_strings":["Iowa State University\u2028"],"raw_orcid":"https://orcid.org/0000-0001-8672-5317","affiliations":[{"raw_affiliation_string":"Iowa State University\u2028","institution_ids":["https://openalex.org/I173911158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":8,"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":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"}},{"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":0.7120000123977661},{"id":"https://openalex.org/keywords/graph-neural-networks","display_name":"graph neural networks","score":0.8289999961853027},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"reinforcement learning","score":0.7360000014305115},{"id":"https://openalex.org/keywords/network-embedding","display_name":"graph embedding","score":0.7509999871253967},{"id":"https://openalex.org/keywords/channel-pruning","display_name":"channel pruning","score":0.8149999976158142},{"id":"https://openalex.org/keywords/model-compression","display_name":"model compression","score":0.7680000066757202},{"id":"https://openalex.org/keywords/flops-reduction","display_name":"FLOPs reduction","score":0.8309999704360962},{"id":"https://openalex.org/keywords/memory-constrained-deployment","display_name":"memory-constrained deployment","score":0.5730000138282776},{"id":"https://openalex.org/keywords/cellular-networks","display_name":"cellular networks","score":0.4399999976158142},{"id":"https://openalex.org/keywords/resnet","display_name":"ResNet","score":0.5899999737739563},{"id":"https://openalex.org/keywords/vgg16","display_name":"VGG16","score":0.9340000152587891},{"id":"https://openalex.org/keywords/mobilenet","display_name":"MobileNet","score":0.953000009059906},{"id":"https://openalex.org/keywords/shufflenet","display_name":"ShuffleNet","score":1.0}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7679368257522583},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.743744432926178},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.706754207611084},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6411746740341187},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5118040442466736},{"id":"https://openalex.org/C199845137","wikidata":"https://www.wikidata.org/wiki/Q145490","display_name":"Network topology","level":2,"score":0.45779305696487427},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.45498669147491455},{"id":"https://openalex.org/C184720557","wikidata":"https://www.wikidata.org/wiki/Q7825049","display_name":"Topology (electrical circuits)","level":2,"score":0.4361088275909424},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.4182116389274597},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.379290372133255},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3760792016983032},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2715005874633789},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.16026565432548523},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.14317670464515686},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11375471949577332},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:3129112952","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2102.03214.pdf","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:3129112952","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2102.03214.pdf","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":35,"referenced_works":["https://openalex.org/W1821462560","https://openalex.org/W2037933327","https://openalex.org/W2058641082","https://openalex.org/W2117539524","https://openalex.org/W2164278908","https://openalex.org/W2173248099","https://openalex.org/W2194775991","https://openalex.org/W2276892413","https://openalex.org/W2300242332","https://openalex.org/W2319920447","https://openalex.org/W2507318699","https://openalex.org/W2515385951","https://openalex.org/W2519887557","https://openalex.org/W2604314403","https://openalex.org/W2612445135","https://openalex.org/W2752037867","https://openalex.org/W2757143157","https://openalex.org/W2798170643","https://openalex.org/W2811124557","https://openalex.org/W2883780447","https://openalex.org/W2886851211","https://openalex.org/W2944779197","https://openalex.org/W2946948417","https://openalex.org/W2948751033","https://openalex.org/W2963125010","https://openalex.org/W2963163009","https://openalex.org/W2963363373","https://openalex.org/W2963993763","https://openalex.org/W2964299589","https://openalex.org/W2970971581","https://openalex.org/W2997768846","https://openalex.org/W3043093753","https://openalex.org/W3093858143","https://openalex.org/W3118608800","https://openalex.org/W3171298560"],"related_works":["https://openalex.org/W7093772913","https://openalex.org/W3214643749","https://openalex.org/W4393303523","https://openalex.org/W7104368479","https://openalex.org/W3107993857","https://openalex.org/W2993389083","https://openalex.org/W3201902125","https://openalex.org/W4385566558","https://openalex.org/W4415428495","https://openalex.org/W2949393176"],"abstract_inverted_index":{"Model":[0],"compression":[1,70,85],"is":[2],"an":[3],"essential":[4],"technique":[5,49],"for":[6,157],"deploying":[7],"deep":[8],"neural":[9,53],"networks":[10,54,130],"(DNNs)":[11],"on":[12,23,28,51,105],"power":[13],"and":[14,26,61,79,92,94,99,119,148,172],"memory-constrained":[15],"resources.":[16],"However,":[17],"existing":[18],"model-compression":[19],"methods":[20,86,147],"often":[21],"rely":[22],"human":[24],"expertise":[25],"focus":[27],"parameters'":[29],"local":[30],"importance,":[31],"ignoring":[32],"the":[33,58,163],"rich":[34],"topology":[35,60],"information":[36],"within":[37],"DNNs.":[38],"In":[39],"this":[40],"paper,":[41],"we":[42],"propose":[43],"a":[44,68,150,167],"novel":[45],"multi-stage":[46],"graph":[47,52],"embedding":[48],"based":[50],"(GNNs)":[55],"to":[56,66,110,135,155],"identify":[57],"DNNs'":[59],"use":[62],"reinforcement":[63],"learning":[64],"(RL)":[65],"find":[67],"suitable":[69],"policy.":[71],"We":[72,101],"performed":[73],"resource-constrained":[74],"(i.e.,":[75],"FLOPs)":[76],"channel":[77],"pruning":[78],"compared":[80],"our":[81,103,125,143,161],"approach":[82],"with":[83],"state-of-the-art":[84,146],"using":[87],"over-parameterized":[88],"DNNs":[89,96],"(e.g.,":[90,97,131],"ResNet":[91,115],"VGG-16)":[93,132],"mobile-friendly":[95,111],"MobileNet":[98],"ShuffleNet).":[100],"evaluated":[102],"method":[104,126,144],"various":[106],"models":[107],"from":[108],"typical":[109],"networks,":[112],"such":[113],"as":[114],"family,":[116],"VGG-16,":[117],"MobileNet-v1/v2,":[118],"ShuffleNet.":[120],"The":[121],"results":[122],"demonstrate":[123],"that":[124],"can":[127],"prune":[128],"dense":[129],"by":[133,153],"up":[134,154,171],"80%":[136],"of":[137],"their":[138],"original":[139],"FLOPs.":[140],"More":[141],"importantly,":[142],"outperformed":[145],"achieved":[149,166],"higher":[151],"accuracy":[152],"1.84%":[156],"ShuffleNet-v1.":[158],"Furthermore,":[159],"following":[160],"approach,":[162],"pruned":[164],"VGG-16":[165],"noticeable":[168],"1.38$\\times$":[169],"speed":[170],"141":[173],"MB":[174],"GPU":[175],"memory":[176],"reduction.":[177]},"counts_by_year":[{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":2}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2021-02-15T00:00:00"}