Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations
收藏国家科技图书文献中心2026-05-09 收录
下载链接：
http://portal.nstl.gov.cn/reportFront/getReportDetailFront.htm?serverId=14&uuid=066e7a777a8e20c26a8e861f230e6666
下载链接
链接失效反馈官方服务：
资源简介：
ThisNISTTrustworthyandResponsibleAIreportdevelopsataxonomyofconceptsanddefinesterminologyinthefieldofadversarialmachinelearning(AML).ThetaxonomyisbuiltonsurveyingtheAMLliteratureandisarrangedinaconceptualhierarchythatincludeskeytypesofMLmethodsandlifecyclestagesofattack,attackergoalsandobjectives,andattackercapabilitiesandknowledgeofthelearningprocess.ThereportalsoprovidescorrespondingmethodsformitigatingandmanagingtheconsequencesofattacksandpointsoutrelevantopenchallengestotakeintoaccountinthelifecycleofAIsystems.TheterminologyusedinthereportisconsistentwiththeliteratureonAMLandiscomplementedbyaglossarythatdefineskeytermsassociatedwiththesecurityofAIsystemsandisintendedtoassistnon-expertreaders.Takentogether,thetaxonomyandterminologyaremeanttoinformotherstandardsandfuturepracticeguidesforassessingandmanagingthesecurityofAIsystems,byestablishingacommonlanguageandunderstandingoftherapidlydevelopingAMLlandscape.
提供机构：
国家标准与技术研究院
