{"id":"https://openalex.org/W2891455536","doi":"https://doi.org/10.18653/v1/d18-1472","title":"Is it Time to Swish? Comparing Deep Learning Activation Functions Across NLP tasks","display_name":"Is it Time to Swish? Comparing Deep Learning Activation Functions Across NLP tasks","publication_year":2018,"publication_date":"2018-01-01","ids":{"openalex":"https://openalex.org/W2891455536","doi":"https://doi.org/10.18653/v1/d18-1472","mag":"2891455536"},"language":"en","primary_location":{"id":"doi:10.18653/v1/d18-1472","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/d18-1472","pdf_url":"https://www.aclweb.org/anthology/D18-1472.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.aclweb.org/anthology/D18-1472.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5053947568","display_name":"Steffen Eger","orcid":"https://orcid.org/0000-0003-4663-8336"},"institutions":[{"id":"https://openalex.org/I31512782","display_name":"Technical University of Darmstadt","ror":"https://ror.org/05n911h24","country_code":"DE","type":"education","lineage":["https://openalex.org/I31512782"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Steffen Eger","raw_affiliation_strings":["Ubiquitous Knowledge Processing Lab (UKP-TUDA) Department of Computer Science Technische Universitt Darmstadt www","Technical University of Darmstadt, Darmstadt, Germany"],"affiliations":[{"raw_affiliation_string":"Ubiquitous Knowledge Processing Lab (UKP-TUDA) Department of Computer Science Technische Universitt Darmstadt www","institution_ids":["https://openalex.org/I31512782"]},{"raw_affiliation_string":"Technical University of Darmstadt, Darmstadt, Germany","institution_ids":["https://openalex.org/I31512782"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055731225","display_name":"Paul Youssef","orcid":"https://orcid.org/0009-0004-4953-4553"},"institutions":[{"id":"https://openalex.org/I31512782","display_name":"Technical University of Darmstadt","ror":"https://ror.org/05n911h24","country_code":"DE","type":"education","lineage":["https://openalex.org/I31512782"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Paul Youssef","raw_affiliation_strings":["Ubiquitous Knowledge Processing Lab (UKP-TUDA) Department of Computer Science Technische Universitt Darmstadt www"],"affiliations":[{"raw_affiliation_string":"Ubiquitous Knowledge Processing Lab (UKP-TUDA) Department of Computer Science Technische Universitt Darmstadt www","institution_ids":["https://openalex.org/I31512782"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5027450194","display_name":"Iryna Gurevych","orcid":"https://orcid.org/0000-0003-2187-7621"},"institutions":[{"id":"https://openalex.org/I31512782","display_name":"Technical University of Darmstadt","ror":"https://ror.org/05n911h24","country_code":"DE","type":"education","lineage":["https://openalex.org/I31512782"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Iryna Gurevych","raw_affiliation_strings":["Ubiquitous Knowledge Processing Lab (UKP-TUDA) Department of Computer Science Technische Universitt Darmstadt www","Technical University of Darmstadt, Darmstadt, Germany"],"affiliations":[{"raw_affiliation_string":"Ubiquitous Knowledge Processing Lab (UKP-TUDA) Department of Computer Science Technische Universitt Darmstadt www","institution_ids":["https://openalex.org/I31512782"]},{"raw_affiliation_string":"Technical University of Darmstadt, Darmstadt, Germany","institution_ids":["https://openalex.org/I31512782"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5053947568"],"corresponding_institution_ids":["https://openalex.org/I31512782"],"apc_list":null,"apc_paid":null,"fwci":1.6923,"has_fulltext":true,"cited_by_count":14,"citation_normalized_percentile":{"value":0.88300675,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"4415","last_page":"4424"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9997000098228455,"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"}},"topics":[{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9997000098228455,"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":0.9993000030517578,"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":0.9973999857902527,"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/activation-function","display_name":"Activation function","score":0.837631106376648},{"id":"https://openalex.org/keywords/sigmoid-function","display_name":"Sigmoid function","score":0.7142077684402466},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6538063287734985},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6382090449333191},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6089849472045898},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.5954392552375793},{"id":"https://openalex.org/keywords/hyperbolic-function","display_name":"Hyperbolic function","score":0.5849559903144836},{"id":"https://openalex.org/keywords/competitor-analysis","display_name":"Competitor analysis","score":0.46938058733940125},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.4335687458515167},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4314756393432617},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4217855632305145},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.41333702206611633},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.41254034638404846},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3933877646923065},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11015275120735168},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.051674842834472656}],"concepts":[{"id":"https://openalex.org/C38365724","wikidata":"https://www.wikidata.org/wiki/Q4677469","display_name":"Activation function","level":3,"score":0.837631106376648},{"id":"https://openalex.org/C81388566","wikidata":"https://www.wikidata.org/wiki/Q526668","display_name":"Sigmoid function","level":3,"score":0.7142077684402466},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6538063287734985},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6382090449333191},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6089849472045898},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.5954392552375793},{"id":"https://openalex.org/C92047909","wikidata":"https://www.wikidata.org/wiki/Q204034","display_name":"Hyperbolic function","level":2,"score":0.5849559903144836},{"id":"https://openalex.org/C127576917","wikidata":"https://www.wikidata.org/wiki/Q624630","display_name":"Competitor analysis","level":2,"score":0.46938058733940125},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.4335687458515167},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4314756393432617},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4217855632305145},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.41333702206611633},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.41254034638404846},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3933877646923065},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11015275120735168},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.051674842834472656},{"id":"https://openalex.org/C78458016","wikidata":"https://www.wikidata.org/wiki/Q840400","display_name":"Evolutionary biology","level":1,"score":0.0},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"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/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.18653/v1/d18-1472","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/d18-1472","pdf_url":"https://www.aclweb.org/anthology/D18-1472.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:1901.02671","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1901.02671","pdf_url":"https://arxiv.org/pdf/1901.02671","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"mag:2891455536","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1901.02671v1","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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},{"id":"pmh:oai:tubiblio.ulb.tu-darmstadt.de:106845","is_oa":false,"landing_page_url":"http://tubiblio.ulb.tu-darmstadt.de/106845/","pdf_url":null,"source":{"id":"https://openalex.org/S4377196390","display_name":"TUbilio (Technical University of Darmstadt)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I31512782","host_organization_name":"Technische Universit\u00e4t Darmstadt","host_organization_lineage":["https://openalex.org/I31512782"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Konferenzver\u00f6ffentlichung"},{"id":"doi:10.48550/arxiv.1901.02671","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1901.02671","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.18653/v1/d18-1472","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/d18-1472","pdf_url":"https://www.aclweb.org/anthology/D18-1472.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G352791218","display_name":null,"funder_award_id":"(BMBF)","funder_id":"https://openalex.org/F4320321114","funder_display_name":"Bundesministerium f\u00fcr Bildung und Forschung"}],"funders":[{"id":"https://openalex.org/F4320321114","display_name":"Bundesministerium f\u00fcr Bildung und Forschung","ror":"https://ror.org/04pz7b180"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2891455536.pdf","grobid_xml":"https://content.openalex.org/works/W2891455536.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W1677182931","https://openalex.org/W1832693441","https://openalex.org/W1921523184","https://openalex.org/W1988115241","https://openalex.org/W2117539524","https://openalex.org/W2156387975","https://openalex.org/W2250861254","https://openalex.org/W2274287116","https://openalex.org/W2279639021","https://openalex.org/W2294059674","https://openalex.org/W2343649478","https://openalex.org/W2402302915","https://openalex.org/W2605035112","https://openalex.org/W2740721704","https://openalex.org/W2752840253","https://openalex.org/W2962753606","https://openalex.org/W2962972936","https://openalex.org/W2963285578","https://openalex.org/W2963454111","https://openalex.org/W2963591087","https://openalex.org/W2963626623","https://openalex.org/W2963918774","https://openalex.org/W2963940534","https://openalex.org/W3118608800"],"related_works":["https://openalex.org/W2963916998","https://openalex.org/W2943971163","https://openalex.org/W3029445693","https://openalex.org/W2947682002","https://openalex.org/W2765833400","https://openalex.org/W3043820487","https://openalex.org/W3042896572","https://openalex.org/W2767072113","https://openalex.org/W2786627679","https://openalex.org/W3195180572","https://openalex.org/W3100141704","https://openalex.org/W3190619293","https://openalex.org/W3110418272","https://openalex.org/W2962953733","https://openalex.org/W2808168001","https://openalex.org/W2785939352","https://openalex.org/W2966253915","https://openalex.org/W3090334579","https://openalex.org/W2894182255","https://openalex.org/W2887282701"],"abstract_inverted_index":{"Activation":[0],"functions":[1,32,46,56,79],"play":[2],"a":[3,88,122,133],"crucial":[4],"role":[5],"in":[6,117],"neural":[7],"networks":[8],"because":[9],"they":[10],"are":[11],"the":[12,19,27,72,99,112,129],"nonlinearities":[13],"which":[14],"have":[15,38],"been":[16,40],"attributed":[17],"to":[18,121],"success":[20],"story":[21],"of":[22,26,76],"deep":[23],"learning.":[24],"One":[25],"currently":[28],"most":[29,50,94],"popular":[30],"activation":[31,55,78,91],"is":[33],"ReLU,":[34],"but":[35],"several":[36],"competitors":[37,67],"recently":[39],"proposed":[41,54],"or":[42],"'discovered',":[43],"including":[44],"LReLU":[45],"and":[47,64,114],"swish.":[48],"While":[49],"works":[51],"compare":[52],"newly":[53],"on":[57,132],"few":[58,66],"tasks":[59],"(usually":[60,68],"from":[61],"image":[62],"classification)":[63],"against":[65],"ReLU),":[69],"we":[70],"perform":[71],"first":[73],"largescale":[74],"comparison":[75],"21":[77],"across":[80,96],"eight":[81],"different":[82],"NLP":[83,135],"tasks.":[84],"We":[85,104],"find":[86],"that":[87,107],"largely":[89],"unknown":[90],"function":[92],"performs":[93],"stably":[95],"all":[97],"tasks,":[98],"so-called":[100],"penalized":[101],"tanh":[102,115],"function.":[103],"also":[105],"show":[106],"it":[108],"can":[109],"successfully":[110],"replace":[111],"sigmoid":[113],"gates":[116],"LSTM":[118],"cells,":[119],"leading":[120],"2":[123],"percentage":[124],"point":[125],"(pp)":[126],"improvement":[127],"over":[128],"standard":[130],"choices":[131],"challenging":[134],"task.":[136]},"counts_by_year":[{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":4},{"year":2020,"cited_by_count":5},{"year":2019,"cited_by_count":1}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-10-10T00:00:00"}