Matches in SemOpenAlex for { <https://semopenalex.org/work/W4283512325> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4283512325 endingPage "117964" @default.
- W4283512325 startingPage "117964" @default.
- W4283512325 abstract "Deep learning is a type of machine learning that adapts a deep hierarchy of concepts. Deep learning classifiers link the most basic version of concepts at the input layer to the most abstract version of concepts at the output layer, also known as a class or label. However, once trained over a finite set of classes, some deep learning models do not have the power to say that a given input does not belong to any of the classes and simply cannot be linked. Correctly invalidating the prediction of unrelated classes is a challenging problem that has been tackled in many ways in the literature. Novelty detection gives deep learning the ability to output “do not know” for novel/unseen classes. Still, no attention has been given to the security aspects of novelty detection. In this paper, we consider the case study of abstraction-based novelty detection and show its weakness against adversarial samples. We show the feasibility of crafting adversarial samples that bypass the novelty detection monitoring and fool the deep learning classifier at the same time. In other words, novelty detection itself ends up as an attack surface. Moreover, we call for further research from a defender’s point of view. We investigate auto-encoders as a plausible defense mechanism and assess its performance." @default.
- W4283512325 created "2022-06-27" @default.
- W4283512325 creator A5043344581 @default.
- W4283512325 creator A5081518084 @default.
- W4283512325 date "2022-11-01" @default.
- W4283512325 modified "2023-10-06" @default.
- W4283512325 title "On the security of deep learning novelty detection" @default.
- W4283512325 cites W1985065696 @default.
- W4283512325 cites W2005289489 @default.
- W4283512325 cites W2053757129 @default.
- W4283512325 cites W2243397390 @default.
- W4283512325 cites W2278186031 @default.
- W4283512325 cites W2795213472 @default.
- W4283512325 cites W2803831897 @default.
- W4283512325 cites W2808237433 @default.
- W4283512325 cites W2925312408 @default.
- W4283512325 cites W2951748425 @default.
- W4283512325 cites W2975763460 @default.
- W4283512325 cites W3005086430 @default.
- W4283512325 cites W3088909400 @default.
- W4283512325 cites W3103145119 @default.
- W4283512325 cites W3158507034 @default.
- W4283512325 cites W3200594578 @default.
- W4283512325 doi "https://doi.org/10.1016/j.eswa.2022.117964" @default.
- W4283512325 hasPublicationYear "2022" @default.
- W4283512325 type Work @default.
- W4283512325 citedByCount "2" @default.
- W4283512325 countsByYear W42835123252022 @default.
- W4283512325 crossrefType "journal-article" @default.
- W4283512325 hasAuthorship W4283512325A5043344581 @default.
- W4283512325 hasAuthorship W4283512325A5081518084 @default.
- W4283512325 hasConcept C101738243 @default.
- W4283512325 hasConcept C108583219 @default.
- W4283512325 hasConcept C119857082 @default.
- W4283512325 hasConcept C138885662 @default.
- W4283512325 hasConcept C154945302 @default.
- W4283512325 hasConcept C27206212 @default.
- W4283512325 hasConcept C2778738651 @default.
- W4283512325 hasConcept C2778924833 @default.
- W4283512325 hasConcept C37736160 @default.
- W4283512325 hasConcept C41008148 @default.
- W4283512325 hasConcept C95623464 @default.
- W4283512325 hasConceptScore W4283512325C101738243 @default.
- W4283512325 hasConceptScore W4283512325C108583219 @default.
- W4283512325 hasConceptScore W4283512325C119857082 @default.
- W4283512325 hasConceptScore W4283512325C138885662 @default.
- W4283512325 hasConceptScore W4283512325C154945302 @default.
- W4283512325 hasConceptScore W4283512325C27206212 @default.
- W4283512325 hasConceptScore W4283512325C2778738651 @default.
- W4283512325 hasConceptScore W4283512325C2778924833 @default.
- W4283512325 hasConceptScore W4283512325C37736160 @default.
- W4283512325 hasConceptScore W4283512325C41008148 @default.
- W4283512325 hasConceptScore W4283512325C95623464 @default.
- W4283512325 hasFunder F4320309999 @default.
- W4283512325 hasFunder F4320336463 @default.
- W4283512325 hasLocation W42835123251 @default.
- W4283512325 hasOpenAccess W4283512325 @default.
- W4283512325 hasPrimaryLocation W42835123251 @default.
- W4283512325 hasRelatedWork W1532481220 @default.
- W4283512325 hasRelatedWork W1939982668 @default.
- W4283512325 hasRelatedWork W2064636555 @default.
- W4283512325 hasRelatedWork W2076090200 @default.
- W4283512325 hasRelatedWork W2081173909 @default.
- W4283512325 hasRelatedWork W2585503716 @default.
- W4283512325 hasRelatedWork W2995944953 @default.
- W4283512325 hasRelatedWork W3025682415 @default.
- W4283512325 hasRelatedWork W4312933423 @default.
- W4283512325 hasRelatedWork W4382317424 @default.
- W4283512325 hasVolume "207" @default.
- W4283512325 isParatext "false" @default.
- W4283512325 isRetracted "false" @default.
- W4283512325 workType "article" @default.