Online Large-Scale Hypothesis Tesng with Corrupted Data

Alves, Victor Benicio Ardilha da Silva

Repositório Institucional da Produção Científica da Marinha do Brasil (RI-MB)

Use este identificador para citar ou linkar para este item: https://www.repositorio.mar.mil.br/handle/ripcmb/847104

Registro completo de metadados

Campo DC	Valor	Idioma
dc.contributor.advisor	Szechtman, Roberto	pt_BR
dc.contributor.advisor	Chen, Louis	pt_BR
dc.contributor.author	Alves, Victor Benicio Ardilha da Silva	-
dc.date.accessioned	2024-09-19T14:20:23Z	-
dc.date.available	2024-09-19T14:20:23Z	-
dc.date.issued	2024	-
dc.identifier.uri	https://www.repositorio.mar.mil.br/handle/ripcmb/847104	-
dc.description	This thesis examines the robustness of the Levels Based On Recent Discovery (LORD) algorithmwhenexposedtocorrupteddata,particularlywithincriticalreal-timeprocessing environments like the Brazilian Navy’s Blue Amazon Management System (SisGAAz). Ourstudyrevealsthatmaintainingtheintegrityofstatisticaltestingiscrucial,mainlywhere decision-makingdependsontheaccuracyofdataanalysisconductedonline. Ourresearchidentifiesandrigorouslyevaluateseffectivemitigationstrategiesagainstprobabilisticdatacorruptionscenarios.Keyfindingshighlighttherobustefficacyof“phantom” rejections and the strategic integration of the LORD algorithm with the online Benjamini andHochberg(BH)algorithm,avariationadaptedfromthetraditionalofflineBHmethod. These approaches, we assert, maintain testing power significantly, even under adversarial manipulations,instillingconfidenceintheireffectiveness. Weproposeacontrolledadversarialsetupinvolvingtwoentities:“Blue,”thedefenderwho aims to make true discoveries, and “Red,” the attacker focused on data corruption. Our analysis investigates several attack scenarios. The first is a singular anticipated attack that manipulatesthefirsttruediscoveryandtraditionallytriggersacascadeeffect,counteredby adjusting the decay rate of each test level to buffer against such disruptions. Additionally, we explore multiple p-value corruption scenarios where strategically placed “phantom” rejections can reclaim compromised testing power, although this strategy faces practical challenges due to the necessity of predicting attack probabilities. Lastly, indiscriminate attacks on any p-value show that integrating the LORD algorithm with the online BH algorithm is exceptionally effective, maintaining the algorithm’s robustness even amidst widespreadcorruption. The thesis concludes that while prevalent algorithms are adequate for handling FDR in trustworthydatascenarios,theireffectivenessdiminishesunderadversarialdatamanipulation, a common issue in real-time data environments. Our findings suggest that enhancing algorithmic robustness against data corruption supports reliability in statistical testing and contributes to broader research and application in adversarial conditions. We propose new avenues for future investigation, such as exploring data corruption impacts on other existing algorithms and developing a “pure” algorithm. This new algorithm could offer a more robustalternativetothecurrentmixedapproach,providingastrongerdefenseagainstdata manipulation.	pt_BR
dc.language.iso	en	pt_BR
dc.publisher	Naval Postgraduate School (NVS)	pt_BR
dc.publisher	Naval Postgraduate School (NVS)	pt_BR
dc.rights	openAccess	pt_BR
dc.subject	False discovery rate	pt_BR
dc.subject	Power	pt_BR
dc.subject	Data corrupon	pt_BR
dc.subject	Cascading eﬀect	pt_BR
dc.title	Online Large-Scale Hypothesis Tesng with Corrupted Data	pt_BR
dc.type	masterThesis	pt_BR
dc.subject.dgpm	Engenharia de produção aplicada à pesquisa operacional e gestão da inovação	pt_BR
Aparece nas coleções:	Engenharia Naval: Coleção de Dissertações

Arquivos associados a este item:

Arquivo	Descrição	Tamanho	Formato
Dissertação - CC Benicio.pdf		8,21 MB	Adobe PDF	Visualizar/Abrir

Mostrar registro simples do item Visualizar estatísticas