{"id":"https://openalex.org/W3013192172","doi":null,"title":"Fast and Stable Adversarial Training through Noise Injection.","display_name":"Fast and Stable Adversarial Training through Noise Injection.","publication_year":2020,"publication_date":"2020-02-24","ids":{"openalex":"https://openalex.org/W3013192172","mag":"3013192172"},"language":"en","primary_location":{"id":"mag:3013192172","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2002.10097v1","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/abs/2002.10097v1","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5028233502","display_name":"Leo Schwinn","orcid":"https://orcid.org/0000-0003-3967-2202","observed_orcids":["https://orcid.org/0000-0003-3967-2202"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Leo Schwinn","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0003-3967-2202","affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5014144494","display_name":"Bjoern M. Eskofier","orcid":"https://orcid.org/0000-0002-0417-0336","observed_orcids":["https://orcid.org/0000-0002-0417-0336"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bjoern M. Eskofier","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0000-0002-0417-0336","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":1,"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":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/T11424","display_name":"Security and Verification in Computing","score":0.0,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10764","display_name":"Privacy-Preserving Technologies in Data","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-training","display_name":"adversarial training","score":0.9990000128746033},{"id":"https://openalex.org/keywords/noise-injection","display_name":"noise injection","score":0.7850000262260437},{"id":"https://openalex.org/keywords/fgsm","display_name":"FGSM","score":0.8849999904632568},{"id":"https://openalex.org/keywords/projected-gradient-descent","display_name":"projected gradient descent","score":0.7680000066757202},{"id":"https://openalex.org/keywords/adversarial-robustness","display_name":"adversarial robustness","score":0.9290000200271606},{"id":"https://openalex.org/keywords/computational-efficiency","display_name":"computational efficiency","score":0.6119999885559082},{"id":"https://openalex.org/keywords/neural-network-robustness","display_name":"neural network robustness","score":0.7369999885559082}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.9297366142272949},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6561071872711182},{"id":"https://openalex.org/C125112378","wikidata":"https://www.wikidata.org/wiki/Q176640","display_name":"Randomness","level":2,"score":0.5555689930915833},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5298095941543579},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5019383430480957},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.4845956563949585},{"id":"https://openalex.org/C150921843","wikidata":"https://www.wikidata.org/wiki/Q1170431","display_name":"Resampling","level":2,"score":0.4455479085445404},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.43826037645339966},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.41887587308883667},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4165147840976715},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.41121962666511536},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3241633176803589},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.24787181615829468},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.09369304776191711},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"mag:3013192172","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2002.10097v1","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:3013192172","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2002.10097v1","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":28,"referenced_works":["https://openalex.org/W1959608418","https://openalex.org/W2057853719","https://openalex.org/W2108598243","https://openalex.org/W2112796928","https://openalex.org/W2124289529","https://openalex.org/W2145339207","https://openalex.org/W2167277498","https://openalex.org/W2460937040","https://openalex.org/W2519091744","https://openalex.org/W2544860310","https://openalex.org/W2620038827","https://openalex.org/W2763421725","https://openalex.org/W2786104118","https://openalex.org/W2810611310","https://openalex.org/W2913848079","https://openalex.org/W2922942510","https://openalex.org/W2943642558","https://openalex.org/W2949979904","https://openalex.org/W2962710014","https://openalex.org/W2963207607","https://openalex.org/W2963485691","https://openalex.org/W2963557656","https://openalex.org/W2964121744","https://openalex.org/W2964153729","https://openalex.org/W2964253222","https://openalex.org/W2996564870","https://openalex.org/W3118608800","https://openalex.org/W3123294050"],"related_works":["https://openalex.org/W3010258799","https://openalex.org/W4221156101","https://openalex.org/W2996564870","https://openalex.org/W3206581298","https://openalex.org/W3038319179","https://openalex.org/W3152982807","https://openalex.org/W7117453395","https://openalex.org/W4415609588","https://openalex.org/W4389104677","https://openalex.org/W2900663597"],"abstract_inverted_index":{"Adversarial":[0],"training":[1,34,62,164,170,188],"is":[2,23,149],"the":[3,10,64,79,84,104,115,126,136,144,147],"most":[4],"successful":[5],"empirical":[6],"method,":[7],"to":[8,78,103,138,168,186],"increase":[9],"robustness":[11,22],"of":[12,83,92,125,146,176],"neural":[13,105],"networks":[14],"against":[15,47],"adversarial":[16,33,38,61,141,163,169],"attacks":[17,39],"yet.":[18],"Unfortunately,":[19],"this":[20,52],"higher":[21,27],"accompanied":[24],"by":[25],"considerably":[26],"computational":[28],"complexity.":[29],"To":[30],"date,":[31],"only":[32],"with":[35,63,161,171],"expensive":[36,67],"multi-step":[37],"like":[40],"Projected":[41],"Gradient":[42,69],"Descent":[43],"(PGD)":[44],"proved":[45],"effective":[46],"equally":[48],"strong":[49],"attacks.":[50],"In":[51],"paper,":[53],"we":[54,74,96,180],"present":[55],"two":[56],"ideas":[57],"that":[58,156],"combined":[59],"enable":[60],"computationally":[65],"less":[66],"Fast":[68],"Sign":[70],"Method":[71],"(FGSM).":[72],"First,":[73],"add":[75,97],"uniform":[76],"noise":[77,157,191],"initial":[80],"data":[81],"point":[82],"FGSM":[85,162],"attack,":[86],"which":[87],"creates":[88],"a":[89,98,120],"wider":[90],"variety":[91],"stronger":[93],"adversaries.":[94],"Further,":[95],"learnable":[99],"regularization":[100],"step":[101],"prior":[102],"network":[106],"called":[107],"Stochastic":[108],"Augmentation":[109],"Layer":[110],"(SAL).":[111],"Inputs":[112],"propagated":[113],"trough":[114],"SAL":[116],"are":[117],"resampled":[118],"from":[119],"Gaussian":[121],"distribution.":[122],"The":[123],"randomness":[124],"resampling":[127],"at":[128],"inference":[129],"time":[130],"makes":[131],"it":[132],"more":[133],"complicated":[134],"for":[135],"attacker":[137],"construct":[139],"an":[140],"example":[142],"since":[143],"outcome":[145],"model":[148],"not":[150],"known":[151],"in":[152,159,184],"advance.":[153],"We":[154],"show":[155,181],"injection":[158,192],"conjunction":[160],"achieves":[165],"comparable":[166],"results":[167,183],"PGD":[172],"while":[173],"being":[174],"orders":[175],"magnitude":[177],"faster.":[178],"Moreover,":[179],"superior":[182],"comparison":[185],"PGD-based":[187],"when":[189],"combining":[190],"and":[193],"SAL.":[194]},"counts_by_year":[{"year":2020,"cited_by_count":1}],"updated_date":"2026-11-03T10:50:19.383937","created_date":"2020-04-03T00:00:00"}