{"id":"https://openalex.org/W4393152577","doi":"https://doi.org/10.1609/aaai.v38i19.30096","title":"From Hope to Safety: Unlearning Biases of Deep Models via Gradient Penalization in Latent Space","display_name":"From Hope to Safety: Unlearning Biases of Deep Models via Gradient Penalization in Latent Space","publication_year":2024,"publication_date":"2024-03-24","ids":{"openalex":"https://openalex.org/W4393152577","doi":"https://doi.org/10.1609/aaai.v38i19.30096"},"language":"en","primary_location":{"id":"doi:10.1609/aaai.v38i19.30096","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v38i19.30096","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/30096/31932","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://ojs.aaai.org/index.php/AAAI/article/download/30096/31932","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5061634795","display_name":"Maximilian Dreyer","orcid":null},"institutions":[{"id":"https://openalex.org/I2800274787","display_name":"Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute","ror":"https://ror.org/02tbr6331","country_code":"DE","type":"facility","lineage":["https://openalex.org/I2800274787","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Maximilian Dreyer","raw_affiliation_strings":["Fraunhofer Heinrich Hertz Institute"],"affiliations":[{"raw_affiliation_string":"Fraunhofer Heinrich Hertz Institute","institution_ids":["https://openalex.org/I2800274787"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059724969","display_name":"Frederik Pahde","orcid":"https://orcid.org/0000-0002-5681-6231"},"institutions":[{"id":"https://openalex.org/I2800274787","display_name":"Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute","ror":"https://ror.org/02tbr6331","country_code":"DE","type":"facility","lineage":["https://openalex.org/I2800274787","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Frederik Pahde","raw_affiliation_strings":["Fraunhofer Heinrich Hertz Institute"],"affiliations":[{"raw_affiliation_string":"Fraunhofer Heinrich Hertz Institute","institution_ids":["https://openalex.org/I2800274787"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075097456","display_name":"Christopher J. Anders","orcid":"https://orcid.org/0000-0003-3295-8486"},"institutions":[{"id":"https://openalex.org/I4577782","display_name":"Technische Universit\u00e4t Berlin","ror":"https://ror.org/03v4gjf40","country_code":"DE","type":"education","lineage":["https://openalex.org/I4577782"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Christopher J. Anders","raw_affiliation_strings":["Technical University of Berlin"],"affiliations":[{"raw_affiliation_string":"Technical University of Berlin","institution_ids":["https://openalex.org/I4577782"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026451495","display_name":"Wojciech Samek","orcid":"https://orcid.org/0000-0002-6283-3265"},"institutions":[{"id":"https://openalex.org/I2800274787","display_name":"Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute","ror":"https://ror.org/02tbr6331","country_code":"DE","type":"facility","lineage":["https://openalex.org/I2800274787","https://openalex.org/I4923324"]},{"id":"https://openalex.org/I4401727010","display_name":"Berlin Institute for the Foundations of Learning and Data","ror":"https://ror.org/05dsfb086","country_code":null,"type":"facility","lineage":["https://openalex.org/I4401727010","https://openalex.org/I4577782","https://openalex.org/I7877124"]},{"id":"https://openalex.org/I4577782","display_name":"Technische Universit\u00e4t Berlin","ror":"https://ror.org/03v4gjf40","country_code":"DE","type":"education","lineage":["https://openalex.org/I4577782"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Wojciech Samek","raw_affiliation_strings":["Fraunhofer Heinrich Hertz Institute\nTechnical University of Berlin\nBIFOLD \u2013 Berlin Institute for the Foundations of Learning and Data"],"affiliations":[{"raw_affiliation_string":"Fraunhofer Heinrich Hertz Institute\nTechnical University of Berlin\nBIFOLD \u2013 Berlin Institute for the Foundations of Learning and Data","institution_ids":["https://openalex.org/I2800274787","https://openalex.org/I4577782","https://openalex.org/I4401727010"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5017608155","display_name":"Sebastian Lapuschkin","orcid":"https://orcid.org/0000-0002-0762-7258"},"institutions":[{"id":"https://openalex.org/I2800274787","display_name":"Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute","ror":"https://ror.org/02tbr6331","country_code":"DE","type":"facility","lineage":["https://openalex.org/I2800274787","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Sebastian Lapuschkin","raw_affiliation_strings":["Fraunhofer Heinrich Hertz Institute"],"affiliations":[{"raw_affiliation_string":"Fraunhofer Heinrich Hertz Institute","institution_ids":["https://openalex.org/I2800274787"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5061634795"],"corresponding_institution_ids":["https://openalex.org/I2800274787"],"apc_list":null,"apc_paid":null,"fwci":0.6085,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.57161981,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":"38","issue":"19","first_page":"21046","last_page":"21054"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.8795999884605408,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.8795999884605408,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.8686000108718872,"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/T10028","display_name":"Topic Modeling","score":0.7526999711990356,"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/space","display_name":"Space (punctuation)","score":0.5978977084159851},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.43627020716667175},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.4243052899837494},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.394228994846344},{"id":"https://openalex.org/keywords/cognitive-psychology","display_name":"Cognitive psychology","score":0.32901453971862793}],"concepts":[{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.5978977084159851},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43627020716667175},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4243052899837494},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.394228994846344},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.32901453971862793},{"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":"doi:10.1609/aaai.v38i19.30096","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v38i19.30096","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/30096/31932","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},{"id":"pmh:oai:publica.fraunhofer.de:publica/475192","is_oa":true,"landing_page_url":"https://publica.fraunhofer.de/handle/publica/475192","pdf_url":null,"source":{"id":"https://openalex.org/S4306400318","display_name":"Fraunhofer-Publica (Fraunhofer-Gesellschaft)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4923324","host_organization_name":"Fraunhofer-Gesellschaft","host_organization_lineage":["https://openalex.org/I4923324"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"conference paper"}],"best_oa_location":{"id":"doi:10.1609/aaai.v38i19.30096","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v38i19.30096","pdf_url":"https://ojs.aaai.org/index.php/AAAI/article/download/30096/31932","source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.8100000023841858,"display_name":"Peace, Justice and strong institutions"}],"awards":[{"id":"https://openalex.org/G1585467943","display_name":null,"funder_award_id":"101093003","funder_id":"https://openalex.org/F4320334322","funder_display_name":"HORIZON EUROPE Framework Programme"},{"id":"https://openalex.org/G208215675","display_name":null,"funder_award_id":"KI-FOR 5363","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G3031800739","display_name":null,"funder_award_id":"BIFOLD","funder_id":"https://openalex.org/F4320321114","funder_display_name":"Bundesministerium f\u00fcr Bildung und Forschung"},{"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"},{"id":"https://openalex.org/G4533892997","display_name":null,"funder_award_id":"01IS18037","funder_id":"https://openalex.org/F4320321114","funder_display_name":"Bundesministerium f\u00fcr Bildung und Forschung"},{"id":"https://openalex.org/G4956428346","display_name":null,"funder_award_id":"Horizon 2020 research and innovatio","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G5219130566","display_name":null,"funder_award_id":"EU Horizon","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G5441272027","display_name":null,"funder_award_id":"965221","funder_id":"https://openalex.org/F4320334322","funder_display_name":"HORIZON EUROPE Framework Programme"},{"id":"https://openalex.org/G6052429835","display_name":null,"funder_award_id":"(DFG)","funder_id":"https://openalex.org/F4320320879","funder_display_name":"Deutsche Forschungsgemeinschaft"},{"id":"https://openalex.org/G6743167992","display_name":null,"funder_award_id":"965221","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G7225624288","display_name":null,"funder_award_id":"This work was","funder_id":"https://openalex.org/F4320321114","funder_display_name":"Bundesministerium f\u00fcr Bildung und Forschung"},{"id":"https://openalex.org/G7257732812","display_name":null,"funder_award_id":"101093003","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G7624340995","display_name":null,"funder_award_id":"01IS18025A","funder_id":"https://openalex.org/F4320321114","funder_display_name":"Bundesministerium f\u00fcr Bildung und Forschung"},{"id":"https://openalex.org/G8318064016","display_name":null,"funder_award_id":"Horizon","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"},{"id":"https://openalex.org/F4320321114","display_name":"Bundesministerium f\u00fcr Bildung und Forschung","ror":"https://ror.org/04pz7b180"},{"id":"https://openalex.org/F4320334322","display_name":"HORIZON EUROPE Framework Programme","ror":null}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4393152577.pdf"},"referenced_works_count":58,"referenced_works":["https://openalex.org/W569478347","https://openalex.org/W1686810756","https://openalex.org/W1787224781","https://openalex.org/W1834627138","https://openalex.org/W1849277567","https://openalex.org/W2011402106","https://openalex.org/W2194775991","https://openalex.org/W2595017202","https://openalex.org/W2765407302","https://openalex.org/W2903408278","https://openalex.org/W2946948417","https://openalex.org/W2955425717","https://openalex.org/W2958514452","https://openalex.org/W2963483561","https://openalex.org/W2963798744","https://openalex.org/W2963946669","https://openalex.org/W2964416181","https://openalex.org/W2975074147","https://openalex.org/W2981565374","https://openalex.org/W3016970897","https://openalex.org/W3035215724","https://openalex.org/W3035989815","https://openalex.org/W3043789969","https://openalex.org/W3048549109","https://openalex.org/W3083185154","https://openalex.org/W3085109610","https://openalex.org/W3091589898","https://openalex.org/W3100711616","https://openalex.org/W3102785203","https://openalex.org/W3114162576","https://openalex.org/W3162746608","https://openalex.org/W3167964457","https://openalex.org/W3191346550","https://openalex.org/W3212197959","https://openalex.org/W4221158223","https://openalex.org/W4239072543","https://openalex.org/W4281637618","https://openalex.org/W4281874990","https://openalex.org/W4282050378","https://openalex.org/W4287115658","https://openalex.org/W4287126366","https://openalex.org/W4287181906","https://openalex.org/W4287239850","https://openalex.org/W4295134803","https://openalex.org/W4311251289","https://openalex.org/W4322759118","https://openalex.org/W4323066538","https://openalex.org/W4360892878","https://openalex.org/W4367694524","https://openalex.org/W4386065384","https://openalex.org/W4386075875","https://openalex.org/W4387225943","https://openalex.org/W6638891565","https://openalex.org/W6676297131","https://openalex.org/W6687483927","https://openalex.org/W6766104482","https://openalex.org/W6776040976","https://openalex.org/W6838961916"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W2382290278","https://openalex.org/W2478288626","https://openalex.org/W4391913857","https://openalex.org/W2350741829","https://openalex.org/W2530322880"],"abstract_inverted_index":{"Deep":[0],"Neural":[1],"Networks":[2],"are":[3,45,144],"prone":[4],"to":[5,15,60,112],"learning":[6],"spurious":[7],"correlations":[8],"embedded":[9],"in":[10,31,114,121],"the":[11,54,62,74,96,127],"training":[12],"data,":[13],"leading":[14],"potentially":[16],"biased":[17],"predictions.":[18],"This":[19],"poses":[20],"risks":[21],"when":[22],"deploying":[23],"these":[24],"models":[25],"for":[26,36,48,70],"high-stake":[27],"decision-making,":[28],"such":[29,106],"as":[30,102,107],"medical":[32],"applications.":[33],"Current":[34],"methods":[35],"post-hoc":[37],"model":[38,71,80],"correction":[39,72],"either":[40],"require":[41],"input-level":[42],"annotations":[43],"which":[44],"only":[46],"possible":[47],"spatially":[49],"localized":[50],"biases,":[51],"or":[52],"augment":[53],"latent":[55],"feature":[56],"space,":[57],"thereby":[58],"hoping":[59],"enforce":[61],"right":[63],"reasons.":[64],"We":[65,117],"present":[66],"a":[67],"novel":[68],"method":[69],"on":[73,126,146],"concept":[75],"level":[76],"that":[77],"explicitly":[78],"reduces":[79],"sensitivity":[81],"towards":[82],"biases":[83,89,120],"via":[84,90],"gradient":[85],"penalization.":[86],"When":[87],"modeling":[88],"Concept":[91],"Activation":[92],"Vectors,":[93],"we":[94],"highlight":[95],"importance":[97],"of":[98],"choosing":[99],"robust":[100],"directions,":[101],"traditional":[103],"regression-based":[104],"approaches":[105],"Support":[108],"Vector":[109],"Machines":[110],"tend":[111],"result":[113],"diverging":[115],"directions.":[116],"effectively":[118],"mitigate":[119],"controlled":[122],"and":[123,132,138,142],"real-world":[124],"settings":[125],"ISIC,":[128],"Bone":[129],"Age,":[130],"ImageNet":[131],"CelebA":[133],"datasets":[134],"using":[135],"VGG,":[136],"ResNet":[137],"EfficientNet":[139],"architectures.":[140],"Code":[141],"Appendix":[143],"available":[145],"https://github.com/frederikpahde/rrclarc.":[147]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":2}],"updated_date":"2026-03-13T16:22:10.518609","created_date":"2025-10-10T00:00:00"}