版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领
文档简介
Federated
Learning姜育刚,马兴军,吴祖煊/2017/04/federated-learning-collaborative.htmlRecap:week10口 CommonTamperingandDeepfakes口 ImageManipulationDetection口 VideoManipulationDetectionThisWeek口 FederatedLearning口 PrivacyinFederatedLearning口 RobustnessinFederatedLearning口 ChallengesandFutureResearchTraditionalMachineLearningDataModelDataandmodelinonesingleplaceTraditionalMachineLearningDataModelWhat
if
we
need
more
data?DataGatheringUsingmultipleGPUsFederatedLearning:Whatisit?Google:FederatedLearning:CollaborativeMachineLearningwithoutCentralizedTrainingDataFederatedLearning:Challenges,Methods,andFutureDirections,/pdf/1908.07873.pdfNextwordpredictiononmobile.FederatedMachineLearning:ConceptandApplications,/pdf/1902.04885.pdfHorizontalFL(横着切):samefeatures,differentsamplesFederatedLearning:TypesVerticalFL(纵着切):samesamples,differentfeaturesFederatedLearning:TypesFederatedMachineLearning:ConceptandApplications,/pdf/1902.04885.pdfFederatedLearning:TypesFederatedTransferLearning:differentsamples,differentfeaturesFederatedMachineLearning:ConceptandApplications,/pdf/1902.04885.pdfCompareDifferentParadigmsFederatedMachineLearning:ConceptandApplications,/pdf/1902.04885.pdfCompareDifferentParadigmshttps:///projects/distributed-learning-and-collaborative-learning-1/overview/SplitLearningvsFederatedLearningFederatedLearningFrameworksHE:homomorphicencryption SS:secretSharingObjectivesandUpdatesinFLGlobalobjectiveLocalobjective:LocalUpdates:GlobalAggregation(e.g.FedAvg):FederatedLearning–MajorChallengesExpensiveCommunicationSystemsHeterogeneityStatisticalHeterogeneityPrivacyandSecurityConcernsFederatedLearning:Challenges,Methods,andFutureDirections,/pdf/1908.07873.pdfFederatedLearning-HorizontalFederatedLearning:Challenges,Methods,andFutureDirections,/pdf/1908.07873.pdfHFLcanfurtherbedividedinto…?PrivacyandSecurityThreatsLyuetal.“Privacyandrobustnessinfederatedlearning:Attacksanddefenses.”TNNLS,2022.SummaryofThreatModelsFLserver(insider)FLparticipants(insider)Eavesdroppers(outsider)Serviceusers(outsider)□InsidervsOutsider □InsiderAttacksByzantine:theworstattacker,knowseverythingaboutthesystem,doesnotobeytheprotocol,sendarbitraryupdates,evencolludewitheachother.Sybil:takingoverthenetworkbysimulatingmanydummyparticipants,out-votethehonestusersSemi-honestvsMaliciousSemi-honestsettingMalicioussettingTraining-timevsTest-timeStealprivatedata,stealmodel,corruptthemodel(trainingtime)Adversarialattack(testtime)SummaryofAttacksExistingattacksagainstserver-basedFLPoisoningAttacksDatapoisoningvsmodel(weight)poisoningDataPoisoningAttacksinTraditionalML□Dirty-labelPoisoningLabelflipping(onlychangelabels)Dirty-labelbackdoor(changeinputsandlabels)Clean-labelPoisoningClean-labelbackdoor(onlychangeinputs)DataPoisoningAttacksinTraditionalMLAsimplepatterncanmakethemodeltomemorizeFLPoisoningAttacks–ModelPoisoningMaincharacteristics:ChangelocalmodelweightsMostlyByzantineattack(attackercandoanythingtotheweights)CanattackByzantine-robustaggregationmechanismssuchasKrumandcoordinate-wisemedianinsteadofweightedaveragingKrum:PrivacyAttacksForeverycommunicationround,localclientshavethechancetoreverseengineerothers’gradients.Fromthereversedgradients,reverseengineer:RepresentationsMembershipPropertiesSensitiveattributesInVFL:featuresPrivacyAttacks–InferenceAttacksDeepmodelsundertheGAN:informationleakagefromcollaborativedeeplearning,CCS2017InferenceclassrepresentationsusingGANsCIFAR-10horseclassReconstructAlice’sfaceimagePrivacyAttacks–InferenceAttacksComprehensiveprivacyanalysisofdeeplearning:Passiveandactivewhite-boxinferenceattacksagainstcentralizedandfederatedlearning,S&P,2019Inferencemembership:Passiveattacks:observeandinference.Activeattacks:influencethetargetmodelinordertoextractmoreinformation.WeaknessofFL:FLcreatesanenvironmentfor(almost)white-boxattacksPrivacyAttacks–InferenceAttacksOtherinferenceattacks:inferringproperties,trainingdata,labels...DeepLeakagefromGradient(DLG)ImprovedDeepLeakagefromGradient(iDLG)…Defenses–PrivacyDefenseHomomorphic
Encryption:RSAEl
GamalPaillier…Homomorphic
properties:Allows
computation
directly
onencrypted
data(“可算不可见”)Needs
to
be
designed
for
eachalgorithmA
side
note:
attacking
encrypted
FL
is
challengingbut
still
possible!Defenses–PrivacyDefense2.
SecureMultipartyComputation(SMC,Yaosharing):SecureML(data-independentofflinephase+fastonlinephase)Offlinemultiplicationtriplets,truncate,sharingCharacteristics:HighlevelprivacyHighcomputationandcommunicationcostYao'sMillionaires'problemProtocolsforSecureComputations,AndrewChi-ChihYao,1982,UCBerkeleyDefenses–PrivacyDefense2.DifferentialPrivacy(DP):TypesofDP:LocalDPCentralizedDPDistributedDPDefenses–PrivacyDefenseDataflowofstatisticsunderLDP2.DifferentialPrivacy(DP):Defenses–PrivacyDefense2.DifferentialPrivacy(DP):TypesoffrequencyestimationDefenses–PrivacyDefense2.DifferentialPrivacy(DP):Real-worldapplications.Vanilla
FLM:ADPmechanismCentralized
DPM:ADPmechanismLocal
DPM:ADPmechanismE:encryptionD:decryptionDistributed
DPDefenses–ByzantineDefenseAlgorithm:Krum(forByzantinerobustness)Setting:nparticipants,fareByzantine,with𝒏≥𝟐𝒇+𝟑Atcommunicationroundt,𝟏 𝟐 𝒏serverreceives{𝜹𝒕,𝜹𝒕,…,𝜹𝒕}foreach𝜹𝒕:𝒊selecttheclosest(L2distance)n-f-2intoset𝑪𝒊compute𝒔𝒄𝒐𝒓𝒆𝜹𝒕=∑𝒊 𝜹∈𝑪𝒊 𝒊𝜹𝒕−𝜹𝜹𝟏 𝒏𝜹𝒌𝒓𝒖𝒎=𝜹∗=argmin{𝒔𝒄𝒐𝒓𝒆𝜹𝒕 ,…,𝒔𝒄𝒐𝒓𝒆𝜹𝒕}updateglobalparameter:𝒘𝒕.𝟏=𝒘𝒕+𝜹𝒌𝒓𝒖𝒎Blanchardetal.“Machinelearningwithadversaries:Byzantinetolerantgradientdescent.”NeurIPS,2017.Defenses–ByzantineDefenseAlgorithm:Krum(forByzantinerobustness)Blanchard
et
al.
“Machine
learning
with
adversaries:
Byzantine
tolerant
gradient
descent.”
NeurIPS,
2017.红色:攻击梯度蓝色:真实梯度黑色:本地梯度黑色曲线:损失函数Defenses–ByzantineDefenseMorerobustaggregationmethods:Multi-Krum=Krum+Averaging=Krumrobustness+increasedconvergencespeedcoordinate-wisemedian,coordinate-wisetrimmedmeanmedianisnotgoodforconvergenceBulyan=Krum+trimmedmedianMedianandgeometric-median(RobustFederatedAggregation)RFA:approximategeometricmedian(notrobusttoByzantineattacks)Defenses–ByzantineDefenseModelpoisoningattackcanbreakKrumandcoordinate-wisemedianAnalyzingfederatedlearningthroughanadversariallens,ICML2019.𝜏/:adversarialtargetclassr:numberofpoisonedsamples𝐷0:cleandata1𝑤?2:estimationoftheglobalparametersReversedgradientsfromthelastround.Defenses–SybilDefenseFromtraditionalML:RejectonNegativeInfluence(RONI)WithacleanvalidationdatasetItrequiresuniformdistributioninnon-IIDsetting,notgood.FoolsGold:Sybilsharethesameobjective,driftsawayfromtheoriginalobjectiveCoreidea:cosinesimilarityFoolsGold:MitigatingSybilsinFederatedLearningPoisoning,/abs/1808.04866Defenses–SybilDefenseDistributedbackdoorattack(DBA)canbypassbothRFAandFoolsGold.DBA:Distributed
Backdoor
Attacks
against
Federated
Learning,
ICLR
2020.
Defenses
-
SummaryDefenseagainstFederatedLearningPoisoning.n:numberofparticipants.RemainingChallengesandFutureResearch□ CurseofdimensionalityLargermodelsaremorevulnerableSharingweights/gradientsmaynotbeagoodidea□ WeaknessesofcurrentattacksGANattackassumestheclassofdataisfromonesingleparticipantDLG/iDLGworkwithsecond-ordergradientmethod(expensive)andsmallminibatch-gradients(B=8)□ Vulnerabilitytofreeriders:pretendtohavedatabutnot.□ WeaknessofCurrentPrivacy-preservingTechniquesSecureaggregationismorevulnerabletopoisoningattacks
温馨提示
- 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
- 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
- 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
- 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
- 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
- 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
- 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。
最新文档
- 烟台理工学院《编译原理》2023-2024学年第一学期期末试卷
- 烟台大学《空间表现技法》2022-2023学年第一学期期末试卷
- 一年级数学计算题专项练习汇编
- 许昌学院《虚拟现实程序设计基础》2021-2022学年第一学期期末试卷
- 许昌学院《软件工程实验》2022-2023学年第一学期期末试卷
- 徐州工程学院《英语演讲与辩论》2023-2024学年第一学期期末试卷
- 徐州工程学院《微机原理与接口技术》2023-2024学年第一学期期末试卷
- 孩子们的成长与发展蓝图计划
- 课外阅读与自主学习建议计划
- 科技展览会的安全管理与保障总结计划
- 铁路客运服务礼仪(第2版)课件 -模块五 重点旅客服务礼仪
- 【MOOC】模拟电子技术基础-华中科技大学 中国大学慕课MOOC答案
- 保护长江同饮一江水共护母亲河主题班会
- 国家开放大学电大本科《工程经济与管理》2023-2024期末试题及答案(试卷号:1141)
- 美国史智慧树知到期末考试答案章节答案2024年东北师范大学
- 《短视频拍摄与制作》课件-3短视频中期拍摄
- 2024年中国华能财务有限责任公司招聘笔试参考题库含答案解析
- 全过程造价咨询服务方案
- ICU镇静镇痛评分表及程序化流程
- 《红军长征》教学反思三篇.doc
- 财务报表分析常用公式汇总
评论
0/150
提交评论