Matches in SemOpenAlex for { <https://semopenalex.org/work/W4293490131> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4293490131 endingPage "112577" @default.
- W4293490131 startingPage "112577" @default.
- W4293490131 abstract "The puzzling phenomenon of adversarial examples continues to attract significant research within the machine learning community. The confirmation that adversarial examples can arise in natural real-life circumstances has but increased the interest. While several methods have been proposed for both generating adversarial examples and defending against them, in this work we focus on a previous serendipitous discovery indicating that they can be considered as chaotic signals. More specifically, it has been recently shown that measures of chaoticity in the input signal can be used to detect adversarial examples efficiently. In this work, we extend that approach in two aspects, leading to significant improvements in detection accuracy as demonstrated by results obtained in experiments with four datasets and using seven different attack methods. • Applying chaos theory to detect adversarial examples, two different contributions. • Lyapunov exponents in the image space and also spatial information are considered. • Adversarial examples are generated using 7 state-of-the-art methods. • Use of common datasets like MNIST, Fashion-MNIST, CIFAR-10 and Tiny-ImageNet. • Three additional adversarial detection methods are trained and compared. • A proper adaptive attack is developed to test the method. • Results are compared for the different approaches in an ablation study." @default.
- W4293490131 created "2022-08-29" @default.
- W4293490131 creator A5040216542 @default.
- W4293490131 creator A5068267966 @default.
- W4293490131 creator A5076163708 @default.
- W4293490131 date "2022-10-01" @default.
- W4293490131 modified "2023-10-18" @default.
- W4293490131 title "Detecting chaos in adversarial examples" @default.
- W4293490131 cites W1965555277 @default.
- W4293490131 cites W1995443851 @default.
- W4293490131 cites W2010187498 @default.
- W4293490131 cites W2152575748 @default.
- W4293490131 cites W2543927648 @default.
- W4293490131 cites W2963564844 @default.
- W4293490131 cites W2963612069 @default.
- W4293490131 cites W3090459527 @default.
- W4293490131 cites W3161366266 @default.
- W4293490131 doi "https://doi.org/10.1016/j.chaos.2022.112577" @default.
- W4293490131 hasPublicationYear "2022" @default.
- W4293490131 type Work @default.
- W4293490131 citedByCount "0" @default.
- W4293490131 crossrefType "journal-article" @default.
- W4293490131 hasAuthorship W4293490131A5040216542 @default.
- W4293490131 hasAuthorship W4293490131A5068267966 @default.
- W4293490131 hasAuthorship W4293490131A5076163708 @default.
- W4293490131 hasConcept C108583219 @default.
- W4293490131 hasConcept C11413529 @default.
- W4293490131 hasConcept C115961682 @default.
- W4293490131 hasConcept C119857082 @default.
- W4293490131 hasConcept C120665830 @default.
- W4293490131 hasConcept C121332964 @default.
- W4293490131 hasConcept C153180895 @default.
- W4293490131 hasConcept C154945302 @default.
- W4293490131 hasConcept C190502265 @default.
- W4293490131 hasConcept C192209626 @default.
- W4293490131 hasConcept C2777052490 @default.
- W4293490131 hasConcept C37736160 @default.
- W4293490131 hasConcept C41008148 @default.
- W4293490131 hasConceptScore W4293490131C108583219 @default.
- W4293490131 hasConceptScore W4293490131C11413529 @default.
- W4293490131 hasConceptScore W4293490131C115961682 @default.
- W4293490131 hasConceptScore W4293490131C119857082 @default.
- W4293490131 hasConceptScore W4293490131C120665830 @default.
- W4293490131 hasConceptScore W4293490131C121332964 @default.
- W4293490131 hasConceptScore W4293490131C153180895 @default.
- W4293490131 hasConceptScore W4293490131C154945302 @default.
- W4293490131 hasConceptScore W4293490131C190502265 @default.
- W4293490131 hasConceptScore W4293490131C192209626 @default.
- W4293490131 hasConceptScore W4293490131C2777052490 @default.
- W4293490131 hasConceptScore W4293490131C37736160 @default.
- W4293490131 hasConceptScore W4293490131C41008148 @default.
- W4293490131 hasLocation W42934901311 @default.
- W4293490131 hasOpenAccess W4293490131 @default.
- W4293490131 hasPrimaryLocation W42934901311 @default.
- W4293490131 hasRelatedWork W2775543769 @default.
- W4293490131 hasRelatedWork W2810865670 @default.
- W4293490131 hasRelatedWork W2908634608 @default.
- W4293490131 hasRelatedWork W3008694380 @default.
- W4293490131 hasRelatedWork W3013699911 @default.
- W4293490131 hasRelatedWork W3026616975 @default.
- W4293490131 hasRelatedWork W3139282803 @default.
- W4293490131 hasRelatedWork W3156786002 @default.
- W4293490131 hasRelatedWork W4220691143 @default.
- W4293490131 hasRelatedWork W4293490131 @default.
- W4293490131 hasVolume "163" @default.
- W4293490131 isParatext "false" @default.
- W4293490131 isRetracted "false" @default.
- W4293490131 workType "article" @default.