{"id":"https://openalex.org/W3047587361","doi":"https://doi.org/10.1145/3340531.3412130","title":"Can Adversarial Weight Perturbations Inject Neural Backdoors","display_name":"Can Adversarial Weight Perturbations Inject Neural Backdoors","publication_year":2020,"publication_date":"2020-10-19","ids":{"openalex":"https://openalex.org/W3047587361","doi":"https://doi.org/10.1145/3340531.3412130","mag":"3047587361"},"language":"en","primary_location":{"id":"doi:10.1145/3340531.3412130","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3340531.3412130","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM International Conference on Information & Knowledge Management","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2008.01761","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5064222284","display_name":"Siddhant Garg","orcid":"https://orcid.org/0009-0004-0219-511X"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Siddhant Garg","raw_affiliation_strings":["Amazon Alexa, Manhattan Beach, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon Alexa, Manhattan Beach, CA, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000966033","display_name":"Adarsh Kumar","orcid":"https://orcid.org/0000-0003-2919-6302"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Adarsh Kumar","raw_affiliation_strings":["Amazon Alexa, Manhattan Beach, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon Alexa, Manhattan Beach, CA, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048064260","display_name":"Vibhor Goel","orcid":null},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Vibhor Goel","raw_affiliation_strings":["University of Wisconsin-Madison, Madison, WI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wisconsin-Madison, Madison, WI, USA","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035905957","display_name":"Yingyu Liang","orcid":"https://orcid.org/0000-0002-3644-774X"},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yingyu Liang","raw_affiliation_strings":["University of Wisconsin-Madison, Madison, WI, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wisconsin-Madison, Madison, WI, USA","institution_ids":["https://openalex.org/I135310074"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":4.4232,"has_fulltext":false,"cited_by_count":58,"citation_normalized_percentile":{"value":0.95467493,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"2029","last_page":"2032"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":1.0,"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":1.0,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.9506000280380249,"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/adversarial-system","display_name":"Adversarial system","score":0.8932328224182129},{"id":"https://openalex.org/keywords/backdoor","display_name":"Backdoor","score":0.8484121561050415},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7177469730377197},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.6309791803359985},{"id":"https://openalex.org/keywords/adversary","display_name":"Adversary","score":0.6076326966285706},{"id":"https://openalex.org/keywords/norm","display_name":"Norm (philosophy)","score":0.5134505033493042},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.5099414587020874},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4900258779525757},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.39896661043167114},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3406357765197754},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.32765549421310425},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1875794529914856},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.11224284768104553}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.8932328224182129},{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.8484121561050415},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7177469730377197},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.6309791803359985},{"id":"https://openalex.org/C41065033","wikidata":"https://www.wikidata.org/wiki/Q2825412","display_name":"Adversary","level":2,"score":0.6076326966285706},{"id":"https://openalex.org/C191795146","wikidata":"https://www.wikidata.org/wiki/Q3878446","display_name":"Norm (philosophy)","level":2,"score":0.5134505033493042},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.5099414587020874},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4900258779525757},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.39896661043167114},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3406357765197754},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.32765549421310425},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1875794529914856},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.11224284768104553},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3340531.3412130","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3340531.3412130","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM International Conference on Information & Knowledge Management","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2008.01761","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2008.01761","pdf_url":"https://arxiv.org/pdf/2008.01761","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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2008.01761","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2008.01761","pdf_url":"https://arxiv.org/pdf/2008.01761","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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W1673923490","https://openalex.org/W1832693441","https://openalex.org/W1945616565","https://openalex.org/W2014902591","https://openalex.org/W2064675550","https://openalex.org/W2163455955","https://openalex.org/W2163605009","https://openalex.org/W2194775991","https://openalex.org/W2535838896","https://openalex.org/W2609368435","https://openalex.org/W2618530766","https://openalex.org/W2752929869","https://openalex.org/W2774423163","https://openalex.org/W2799194071","https://openalex.org/W2807363941","https://openalex.org/W2810065831","https://openalex.org/W2896457183","https://openalex.org/W2898759955","https://openalex.org/W2903356604","https://openalex.org/W2934843808","https://openalex.org/W2942091739","https://openalex.org/W2952186591","https://openalex.org/W2962718684","https://openalex.org/W2963207607","https://openalex.org/W2963341956","https://openalex.org/W2963422767","https://openalex.org/W2963771448","https://openalex.org/W2963857521","https://openalex.org/W2964041528","https://openalex.org/W2964153729","https://openalex.org/W2998432370","https://openalex.org/W3013520104","https://openalex.org/W3014153848","https://openalex.org/W3016622506","https://openalex.org/W3035367371","https://openalex.org/W3041107652","https://openalex.org/W3098772125","https://openalex.org/W3104423855","https://openalex.org/W3118608800","https://openalex.org/W4287725452","https://openalex.org/W4288104702","https://openalex.org/W4294506858","https://openalex.org/W4297687186"],"related_works":["https://openalex.org/W4320018150","https://openalex.org/W4328053081","https://openalex.org/W4239582170","https://openalex.org/W2918664383","https://openalex.org/W106056076","https://openalex.org/W4320855730","https://openalex.org/W2135200719","https://openalex.org/W2950183588","https://openalex.org/W3080754722","https://openalex.org/W3093978547"],"abstract_inverted_index":{"Adversarial":[0],"machine":[1],"learning":[2],"has":[3,19],"exposed":[4],"several":[5,177],"security":[6,69],"hazards":[7],"of":[8,15,50,56,71,110,144],"neural":[9],"models.":[10,75],"Thus":[11],"far,":[12],"the":[13,26,48,54,87,96,100,108,126,142,145,149],"concept":[14],"an":[16,111,122],"\"adversarial":[17,51],"perturbation\"":[18],"exclusively":[20],"been":[21],"used":[22],"with":[23,165],"reference":[24],"to":[25,30,41,53,60,81,95,118],"input":[27],"space":[28,55],"referring":[29],"a":[31,38,68,78,83,90,104,138,166],"small,":[32],"imperceptible":[33],"change":[34,171],"which":[35,66],"can":[36,161],"cause":[37],"ML":[39],"model":[40,57,88,128,135,147,173],"err.":[42],"In":[43],"this":[44],"work":[45],"we":[46,113],"extend":[47],"idea":[49],"perturbations\"":[52],"weights,":[58],"specifically":[59],"inject":[61],"backdoors":[62,160],"in":[63,134,172],"trained":[64,74],"DNNs,":[65],"exposes":[67],"risk":[70],"publicly":[72],"available":[73],"Here,":[76],"injecting":[77],"backdoor":[79],"refers":[80],"obtaining":[82],"desired":[84,150],"outcome":[85],"from":[86],"when":[89],"trigger":[91,151],"pattern":[92],"is":[93],"added":[94],"input,":[97],"while":[98],"retaining":[99],"original":[101,127,146],"predictions":[102,143],"on":[103,141],"non-triggered":[105],"input.":[106],"From":[107],"perspective":[109],"adversary,":[112],"characterize":[114],"these":[115],"adversarial":[116,132],"perturbations":[117,133],"be":[119,162],"constrained":[120],"within":[121],"\u2113\u221e":[123],"norm":[124],"around":[125],"weights.":[129],"We":[130],"introduce":[131],"weights":[136],"using":[137],"composite":[139],"loss":[140],"and":[148,179],"through":[152],"projected":[153],"gradient":[154],"descent.":[155],"Our":[156],"results":[157],"show":[158],"that":[159],"successfully":[163],"injected":[164],"very":[167],"small":[168],"average":[169],"relative":[170],"weight":[174],"values":[175],"for":[176],"CV":[178],"NLP":[180],"applications.":[181]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":11},{"year":2023,"cited_by_count":9},{"year":2022,"cited_by_count":9},{"year":2021,"cited_by_count":16}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-10T00:00:00"}