{"id":"https://openalex.org/W3171219085","doi":"https://doi.org/10.48550/arxiv.2106.05657","title":"Deep neural network loses attention to adversarial images","display_name":"Deep neural network loses attention to adversarial images","publication_year":2021,"publication_date":"2021-06-10","ids":{"openalex":"https://openalex.org/W3171219085","doi":"https://doi.org/10.48550/arxiv.2106.05657","mag":"3171219085"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2106.05657","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.05657","pdf_url":"https://arxiv.org/pdf/2106.05657","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2106.05657","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5001629940","display_name":"Shashank Kotyan","orcid":"https://orcid.org/0000-0003-1235-3316","observed_orcids":["https://orcid.org/0000-0003-1235-3316"]},"institutions":[{"id":"https://openalex.org/I135598925","display_name":"Kyushu University","ror":"https://ror.org/00p4k0j84","country_code":"JP","type":"education","lineage":["https://openalex.org/I135598925"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Kotyan, Shashank","raw_affiliation_strings":["Kyushu University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kyushu University","institution_ids":["https://openalex.org/I135598925"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5028815763","display_name":"Danilo Vasconcellos Vargas","orcid":"https://orcid.org/0000-0001-7442-1279","observed_orcids":["https://orcid.org/0000-0001-7442-1279"]},"institutions":[{"id":"https://openalex.org/I135598925","display_name":"Kyushu University","ror":"https://ror.org/00p4k0j84","country_code":"JP","type":"education","lineage":["https://openalex.org/I135598925"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Vargas, Danilo Vasconcellos","raw_affiliation_strings":["Kyushu University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Kyushu University","institution_ids":["https://openalex.org/I135598925"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I135598925"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9994999766349792,"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":0.9994999766349792,"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/T12357","display_name":"Digital Media Forensic Detection","score":9.999999747378752e-05,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.0,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/adversarial-examples","display_name":"adversarial examples","score":0.8330000042915344},{"id":"https://openalex.org/keywords/adversarial-attacks","display_name":"adversarial attacks","score":0.6629999876022339},{"id":"https://openalex.org/keywords/image-classification","display_name":"image classification","score":0.6290000081062317},{"id":"https://openalex.org/keywords/saliency-maps","display_name":"saliency maps","score":0.8859999775886536},{"id":"https://openalex.org/keywords/grad-cam","display_name":"Grad-CAM","score":0.9760000109672546},{"id":"https://openalex.org/keywords/attention-maps","display_name":"attention maps","score":0.7609999775886536},{"id":"https://openalex.org/keywords/projected-gradient-descent-attacks","display_name":"projected gradient descent attacks","score":0.9290000200271606},{"id":"https://openalex.org/keywords/neural-network-robustness","display_name":"neural network robustness","score":0.7559999823570251},{"id":"https://openalex.org/keywords/interpretability","display_name":"interpretability","score":0.4189999997615814}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.8823851943016052},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.6377226710319519},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5897329449653625},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5209394693374634},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5149977207183838},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3447996973991394},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.32689395546913147}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:2106.05657","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.05657","pdf_url":"https://arxiv.org/pdf/2106.05657","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2106.05657","doi":"https://doi.org/10.48550/arxiv.2106.05657","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2106.05657","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},{"id":"mag:3198478951","doi":null,"is_oa":false,"landing_page_url":"http://ceur-ws.org/Vol-2916/paper_21.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306419999","display_name":"International Joint Conference on Artificial Intelligence","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":"International Joint Conference on Artificial Intelligence","raw_type":null}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2106.05657","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.05657","pdf_url":"https://arxiv.org/pdf/2106.05657","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G2920157438","display_name":null,"funder_award_id":"JP20241216","funder_id":"https://openalex.org/F4320334764","funder_display_name":"Japan Society for the Promotion of Science"}],"funders":[{"id":"https://openalex.org/F4320334764","display_name":"Japan Society for the Promotion of Science","ror":"https://ror.org/00hhkn466"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3171219085.pdf","grobid_xml":"https://content.openalex.org/works/W3171219085.grobid-xml"},"referenced_works_count":15,"referenced_works":["https://openalex.org/W1932198206","https://openalex.org/W1945616565","https://openalex.org/W2123045220","https://openalex.org/W2169393322","https://openalex.org/W2512416592","https://openalex.org/W2620038827","https://openalex.org/W2785887350","https://openalex.org/W2787708942","https://openalex.org/W2794002979","https://openalex.org/W2804700615","https://openalex.org/W2962858109","https://openalex.org/W2963207607","https://openalex.org/W2964253222","https://openalex.org/W3115790835","https://openalex.org/W4294517337"],"related_works":["https://openalex.org/W3207081511","https://openalex.org/W4294831696","https://openalex.org/W4313349685","https://openalex.org/W4296441440","https://openalex.org/W4400285264","https://openalex.org/W2953834312","https://openalex.org/W4352981715","https://openalex.org/W3115790835","https://openalex.org/W7221112964","https://openalex.org/W4312120836"],"abstract_inverted_index":{"Adversarial":[0],"algorithms":[1,17,37],"have":[2],"shown":[3],"to":[4,74,124,143,224],"be":[5,72],"effective":[6],"against":[7,195,200],"neural":[8,79,121,165,228],"networks":[9],"for":[10,26,169],"a":[11],"variety":[12],"of":[13,67,85,100,106,119,133,211,217],"tasks.":[14],"Some":[15],"adversarial":[16,51,69,107,114,218],"perturb":[18,38],"all":[19],"the":[20,23,27,65,78,90,117,120,131,140,147,152,164,170,180,212,215,222],"pixels":[21,40,137,158],"in":[22,31,130],"image":[24,28,32],"minimally":[25],"classification":[29],"task":[30],"classification.":[33],"In":[34],"contrast,":[35],"some":[36,192,196],"few":[39,103],"strongly.":[41],"However,":[42],"very":[43],"little":[44],"information":[45],"is":[46],"available":[47],"regarding":[48],"why":[49,191],"these":[50,68],"samples":[52,70,115,219],"so":[53,159],"diverse":[54],"from":[55,149],"each":[56],"other":[57,201],"exist.":[58],"Recently,":[59],"Vargas":[60],"et":[61],"al.":[62],"showed":[63],"that":[64,129,160,176,205],"existence":[66,213],"might":[71],"due":[73],"conflicting":[75,86],"saliency":[76,87,181],"within":[77],"network.":[80],"We":[81,109,127,173,203],"test":[82],"this":[83,206],"hypothesis":[84],"by":[88],"analysing":[89],"Saliency":[91],"Maps":[92,98],"(SM)":[93],"and":[94,102,183,214,220],"Gradient-weighted":[95],"Class":[96],"Activation":[97],"(Grad-CAM)":[99],"original":[101,125],"different":[104,113],"types":[105],"samples.":[108,126],"also":[110,174],"analyse":[111],"how":[112],"distort":[116],"attention":[118,142,148,168],"network":[122,141,166],"compared":[123],"show":[128,175],"case":[132],"Pixel":[134],"Attack,":[135],"perturbed":[136],"either":[138],"calls":[139],"themselves":[144],"or":[145],"divert":[146],"them.":[150],"Simultaneously,":[151],"Projected":[153],"Gradient":[154],"Descent":[155],"Attack":[156],"perturbs":[157],"intermediate":[161],"layers":[162],"inside":[163],"lose":[167],"correct":[171],"class.":[172],"both":[177],"attacks":[178,197],"affect":[179],"map":[182],"activation":[184],"maps":[185],"differently.":[186],"Thus,":[187],"shedding":[188],"light":[189],"on":[190],"defences":[193],"successful":[194],"remain":[198],"vulnerable":[199],"attacks.":[202],"hope":[204],"analysis":[207],"will":[208],"improve":[209],"understanding":[210],"effect":[216],"enable":[221],"community":[223],"develop":[225],"more":[226],"robust":[227],"networks.":[229]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2021-06-22T00:00:00"}