{"id":"https://openalex.org/W2962662229","doi":null,"title":"On Activation Function Coresets for Network Pruning","display_name":"On Activation Function Coresets for Network Pruning","publication_year":2019,"publication_date":"2019-07-09","ids":{"openalex":"https://openalex.org/W2962662229","mag":"2962662229"},"language":"en","primary_location":{"id":"mag:2962662229","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1907.04018.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/1907.04018.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5020058163","display_name":"Ben Mussay","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ben Mussay","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018283928","display_name":"Samson Zhou","orcid":"https://orcid.org/0000-0001-8288-5698","observed_orcids":["https://orcid.org/0000-0001-8288-5698"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Samson Zhou","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0001-8288-5698","affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053736722","display_name":"Vladimir Braverman","orcid":"https://orcid.org/0000-0001-7709-8753","observed_orcids":["https://orcid.org/0000-0001-7709-8753","https://orcid.org/0009-0008-9727-357X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vladimir Braverman","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0001-7709-8753","affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5082255004","display_name":"Dan Feldman","orcid":"https://orcid.org/0000-0002-7700-9711","observed_orcids":["https://orcid.org/0000-0002-7700-9711"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dan Feldman","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-7700-9711","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":4,"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":0.9717000126838684,"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":0.9717000126838684,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.011800000444054604,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.006000000052154064,"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/coresets","display_name":"coresets","score":0.6050000190734863},{"id":"https://openalex.org/keywords/network-pruning","display_name":"network pruning","score":0.8050000071525574},{"id":"https://openalex.org/keywords/model-compression","display_name":"model compression","score":0.9539999961853027},{"id":"https://openalex.org/keywords/activation-functions","display_name":"activation functions","score":0.8539999723434448},{"id":"https://openalex.org/keywords/neuron-pruning","display_name":"neuron pruning","score":0.7519999742507935},{"id":"https://openalex.org/keywords/relu","display_name":"ReLU","score":0.6570000052452087},{"id":"https://openalex.org/keywords/sigmoid","display_name":"sigmoid","score":0.9520000219345093},{"id":"https://openalex.org/keywords/lenet","display_name":"LeNet","score":0.5669999718666077},{"id":"https://openalex.org/keywords/mnist","display_name":"MNIST","score":0.7440000176429749}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7751353979110718},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.7177506685256958},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.532450795173645},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5302029848098755},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.5181973576545715},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.5150822997093201},{"id":"https://openalex.org/C81388566","wikidata":"https://www.wikidata.org/wiki/Q526668","display_name":"Sigmoid function","level":3,"score":0.5044986009597778},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.48523736000061035},{"id":"https://openalex.org/C38365724","wikidata":"https://www.wikidata.org/wiki/Q4677469","display_name":"Activation function","level":3,"score":0.4731903076171875},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46790748834609985},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.43006110191345215},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.42586031556129456},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.42570361495018005},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3398948013782501},{"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/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/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C78458016","wikidata":"https://www.wikidata.org/wiki/Q840400","display_name":"Evolutionary biology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:2962662229","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1907.04018.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:2962662229","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1907.04018.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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W174068279","https://openalex.org/W566555209","https://openalex.org/W1586313164","https://openalex.org/W1686810756","https://openalex.org/W2029538739","https://openalex.org/W2112796928","https://openalex.org/W2114766824","https://openalex.org/W2134273960","https://openalex.org/W2167215970","https://openalex.org/W2169630701","https://openalex.org/W2230030897","https://openalex.org/W2618305643","https://openalex.org/W2767827820","https://openalex.org/W2789144738","https://openalex.org/W2807299122","https://openalex.org/W2883070812","https://openalex.org/W2899748887","https://openalex.org/W2915589364","https://openalex.org/W2962950822","https://openalex.org/W2963094099","https://openalex.org/W2963145730","https://openalex.org/W2963363373","https://openalex.org/W2963674932","https://openalex.org/W2964153729","https://openalex.org/W2964299589"],"related_works":["https://openalex.org/W2996319422","https://openalex.org/W7133219626","https://openalex.org/W3173173537","https://openalex.org/W2797713747","https://openalex.org/W2990832304","https://openalex.org/W4391306961","https://openalex.org/W3005165565","https://openalex.org/W2949393176","https://openalex.org/W3163244872","https://openalex.org/W3199754351"],"abstract_inverted_index":{"Model":[0],"compression":[1,33,49,152,242],"provides":[2,238],"a":[3,53,82,84,103,110,164,185,200],"means":[4],"to":[5,98,148,240],"efficiently":[6],"deploy":[7],"deep":[8,43],"neural":[9,106,113],"networks":[10],"(DNNs)":[11],"on":[12,60,117,243,247],"devices":[13],"that":[14,91,161,189,203],"limited":[15],"computation":[16],"resources":[17],"and":[18,25,66,129,195,207,229,249],"tight":[19],"power":[20],"budgets,":[21],"such":[22],"as":[23],"mobile":[24],"IoT":[26],"(Internet":[27],"of":[28,36,88,102,112,120,145,170,174,187,192,220],"Things)":[29],"devices.":[30],"Consequently,":[31],"model":[32,48,151],"is":[34,127,168,183],"one":[35],"the":[37,46,89,94,100,118,121,124,143,154,171,175,213,218,244,252],"most":[38],"critical":[39],"topics":[40],"in":[41],"modern":[42],"learning.":[44],"Typically,":[45],"state-of-the-art":[47],"methods":[50,222],"suffer":[51],"from":[52,184],"big":[54],"limitation:":[55],"they":[56],"are":[57],"only":[58,131,133],"based":[59,116],"heuristics":[61],"rather":[62],"than":[63],"theoretical":[64],"foundation":[65],"thus":[67],"offer":[68],"no":[69],"worst-case":[70],"guarantees.":[71],"To":[72],"bridge":[73],"this":[74],"gap,":[75],"Baykal":[76],"et.":[77],"al.":[78],"[2018a]":[79],"suggested":[80],"using":[81],"coreset,":[83],"small":[85],"weighted":[86],"subset":[87],"data":[90,96,176],"provably":[92],"approximates":[93],"original":[95],"set,":[97],"sparsify":[99],"parameters":[101,115],"trained":[104],"fully-connected":[105],"network":[107,114,232],"by":[108],"sampling":[109,125],"number":[111],"importance":[119],"data.":[122],"However,":[123],"procedure":[126],"data-dependent":[128],"can":[130],"be":[132,134],"performed":[135],"after":[136],"an":[137],"expensive":[138],"training":[139],"phase.":[140],"\r\nWe":[141],"propose":[142],"use":[144],"data-independent":[146],"coresets":[147,206],"perform":[149],"provable":[150],"without":[153],"need":[155],"for":[156,177,212,226],"training.":[157],"We":[158,197,216],"first":[159],"prove":[160],"there":[162],"exists":[163],"coreset":[165],"whose":[166,180],"size":[167,173],"independent":[169],"input":[172],"any":[178],"neuron":[179,210],"activation":[181],"function":[182],"family":[186],"functions":[188],"includes":[190],"variants":[191],"ReLU,":[193],"sigmoid":[194],"others.":[196],"then":[198],"provide":[199],"compression-based":[201],"algorithm":[202],"constructs":[204],"these":[205],"explicitly":[208],"applies":[209],"pruning":[211],"underlying":[214],"model.":[215],"demonstrate":[217],"effectiveness":[219],"our":[221,236],"with":[223],"experimental":[224],"evaluations":[225],"both":[227],"synthetic":[228],"real-world":[230],"benchmark":[231],"compression.":[233],"In":[234],"particular,":[235],"framework":[237],"up":[239],"90%":[241],"LeNet-300-100":[245],"architecture":[246],"MNIST":[248],"actually":[250],"improves":[251],"accuracy.":[253]},"counts_by_year":[{"year":2020,"cited_by_count":3},{"year":2019,"cited_by_count":1}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2019-07-23T00:00:00"}