Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891300960> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2891300960 abstract "Many state-of-the-art machine learning models such as deep neural networks have recently shown to be vulnerable to adversarial perturbations, especially in classification tasks. Motivated by adversarial machine learning, in this paper we investigate the robustness of sparse regression models with strongly correlated covariates to adversarially designed measurement noises. Specifically, we consider the family of ordered weighted l 1 (OWL) regularized regression methods and study the case of OSCAR (octagonal shrinkage clustering algorithm for regression) in the adversarial setting. Under a norm-bounded threat model, we formulate the process of finding a maximally disruptive noise for OWL-regularized regression as an optimization problem and illustrate the steps towards finding such a noise in the case of OSCAR. Experimental results demonstrate that the regression performance of grouping strongly correlated features can be severely degraded under our adversarial setting, even when the noise budget is significantly smaller than the ground-truth signals." @default.
- W2891300960 created "2018-09-27" @default.
- W2891300960 creator A5002976916 @default.
- W2891300960 creator A5017540117 @default.
- W2891300960 creator A5050344371 @default.
- W2891300960 date "2018-11-01" @default.
- W2891300960 modified "2023-09-25" @default.
- W2891300960 title "IS ORDERED WEIGHTED ℓ<inf>1</inf> REGULARIZED REGRESSION ROBUST TO ADVERSARIAL PERTURBATION? A CASE STUDY ON OSCAR" @default.
- W2891300960 cites W2012653948 @default.
- W2891300960 cites W2013990204 @default.
- W2891300960 cites W2099471712 @default.
- W2891300960 cites W2100556411 @default.
- W2891300960 cites W2135046866 @default.
- W2891300960 cites W2746600820 @default.
- W2891300960 cites W2773446523 @default.
- W2891300960 cites W2887603965 @default.
- W2891300960 cites W2962756933 @default.
- W2891300960 cites W2962995403 @default.
- W2891300960 cites W2963207607 @default.
- W2891300960 cites W2963243330 @default.
- W2891300960 cites W2963389226 @default.
- W2891300960 cites W2963612069 @default.
- W2891300960 cites W2963696951 @default.
- W2891300960 cites W2963857521 @default.
- W2891300960 cites W2964153729 @default.
- W2891300960 cites W2964205597 @default.
- W2891300960 cites W9657784 @default.
- W2891300960 doi "https://doi.org/10.1109/globalsip.2018.8646623" @default.
- W2891300960 hasPublicationYear "2018" @default.
- W2891300960 type Work @default.
- W2891300960 sameAs 2891300960 @default.
- W2891300960 citedByCount "0" @default.
- W2891300960 crossrefType "proceedings-article" @default.
- W2891300960 hasAuthorship W2891300960A5002976916 @default.
- W2891300960 hasAuthorship W2891300960A5017540117 @default.
- W2891300960 hasAuthorship W2891300960A5050344371 @default.
- W2891300960 hasConcept C104317684 @default.
- W2891300960 hasConcept C105795698 @default.
- W2891300960 hasConcept C11413529 @default.
- W2891300960 hasConcept C115961682 @default.
- W2891300960 hasConcept C119043178 @default.
- W2891300960 hasConcept C119857082 @default.
- W2891300960 hasConcept C134306372 @default.
- W2891300960 hasConcept C146849305 @default.
- W2891300960 hasConcept C152877465 @default.
- W2891300960 hasConcept C154945302 @default.
- W2891300960 hasConcept C185592680 @default.
- W2891300960 hasConcept C33923547 @default.
- W2891300960 hasConcept C34388435 @default.
- W2891300960 hasConcept C37736160 @default.
- W2891300960 hasConcept C41008148 @default.
- W2891300960 hasConcept C55493867 @default.
- W2891300960 hasConcept C63479239 @default.
- W2891300960 hasConcept C73555534 @default.
- W2891300960 hasConcept C83546350 @default.
- W2891300960 hasConcept C99498987 @default.
- W2891300960 hasConceptScore W2891300960C104317684 @default.
- W2891300960 hasConceptScore W2891300960C105795698 @default.
- W2891300960 hasConceptScore W2891300960C11413529 @default.
- W2891300960 hasConceptScore W2891300960C115961682 @default.
- W2891300960 hasConceptScore W2891300960C119043178 @default.
- W2891300960 hasConceptScore W2891300960C119857082 @default.
- W2891300960 hasConceptScore W2891300960C134306372 @default.
- W2891300960 hasConceptScore W2891300960C146849305 @default.
- W2891300960 hasConceptScore W2891300960C152877465 @default.
- W2891300960 hasConceptScore W2891300960C154945302 @default.
- W2891300960 hasConceptScore W2891300960C185592680 @default.
- W2891300960 hasConceptScore W2891300960C33923547 @default.
- W2891300960 hasConceptScore W2891300960C34388435 @default.
- W2891300960 hasConceptScore W2891300960C37736160 @default.
- W2891300960 hasConceptScore W2891300960C41008148 @default.
- W2891300960 hasConceptScore W2891300960C55493867 @default.
- W2891300960 hasConceptScore W2891300960C63479239 @default.
- W2891300960 hasConceptScore W2891300960C73555534 @default.
- W2891300960 hasConceptScore W2891300960C83546350 @default.
- W2891300960 hasConceptScore W2891300960C99498987 @default.
- W2891300960 hasLocation W28913009601 @default.
- W2891300960 hasOpenAccess W2891300960 @default.
- W2891300960 hasPrimaryLocation W28913009601 @default.
- W2891300960 hasRelatedWork W2610321374 @default.
- W2891300960 hasRelatedWork W2996955411 @default.
- W2891300960 hasRelatedWork W3164178104 @default.
- W2891300960 hasRelatedWork W3164563754 @default.
- W2891300960 hasRelatedWork W3166842523 @default.
- W2891300960 hasRelatedWork W3197436093 @default.
- W2891300960 hasRelatedWork W4226129006 @default.
- W2891300960 hasRelatedWork W4282972241 @default.
- W2891300960 hasRelatedWork W4286988501 @default.
- W2891300960 hasRelatedWork W4287702505 @default.
- W2891300960 isParatext "false" @default.
- W2891300960 isRetracted "false" @default.
- W2891300960 magId "2891300960" @default.
- W2891300960 workType "article" @default.