Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950147665> ?p ?o ?g. }
- W2950147665 abstract "The reliance on deep learning algorithms has grown significantly in recent years. Yet, these models are highly vulnerable to adversarial attacks, which introduce visually imperceptible perturbations into testing data to induce misclassifications. The literature has proposed several methods to combat such adversarial attacks, but each method either fails at high perturbation values, requires excessive computing power, or both. This letter proposes a computationally efficient method for defending the Fast Gradient Sign (FGS) adversarial attack by simultaneously denoising and compressing data. Specifically, our proposed defense relies on training a fully connected multi-layer Denoising Autoencoder (DAE) and using its encoder as a defense against the adversarial attack. Our results show that using this dimensionality reduction scheme is not only highly effective in mitigating the effect of the FGS attack in multiple threat models, but it also provides a 2.43x speedup in comparison to defense strategies providing similar robustness against the same attack." @default.
- W2950147665 created "2019-06-27" @default.
- W2950147665 creator A5037744622 @default.
- W2950147665 creator A5051018107 @default.
- W2950147665 creator A5084897218 @default.
- W2950147665 date "2019-06-13" @default.
- W2950147665 modified "2023-09-27" @default.
- W2950147665 title "A Computationally Efficient Method for Defending Adversarial Deep Learning Attacks" @default.
- W2950147665 cites W1883420340 @default.
- W2950147665 cites W2007562169 @default.
- W2950147665 cites W2095577883 @default.
- W2950147665 cites W2408141691 @default.
- W2950147665 cites W2528914598 @default.
- W2950147665 cites W2590523583 @default.
- W2950147665 cites W2618043096 @default.
- W2950147665 cites W2747237166 @default.
- W2950147665 cites W2750384547 @default.
- W2950147665 cites W2768899812 @default.
- W2950147665 cites W2784285928 @default.
- W2950147665 cites W2805366091 @default.
- W2950147665 cites W2952678275 @default.
- W2950147665 cites W2963564844 @default.
- W2950147665 cites W2963895533 @default.
- W2950147665 cites W2964121744 @default.
- W2950147665 hasPublicationYear "2019" @default.
- W2950147665 type Work @default.
- W2950147665 sameAs 2950147665 @default.
- W2950147665 citedByCount "1" @default.
- W2950147665 countsByYear W29501476652020 @default.
- W2950147665 crossrefType "posted-content" @default.
- W2950147665 hasAuthorship W2950147665A5037744622 @default.
- W2950147665 hasAuthorship W2950147665A5051018107 @default.
- W2950147665 hasAuthorship W2950147665A5084897218 @default.
- W2950147665 hasConcept C101738243 @default.
- W2950147665 hasConcept C104317684 @default.
- W2950147665 hasConcept C108583219 @default.
- W2950147665 hasConcept C111030470 @default.
- W2950147665 hasConcept C11413529 @default.
- W2950147665 hasConcept C119857082 @default.
- W2950147665 hasConcept C134306372 @default.
- W2950147665 hasConcept C139676723 @default.
- W2950147665 hasConcept C154945302 @default.
- W2950147665 hasConcept C173608175 @default.
- W2950147665 hasConcept C185592680 @default.
- W2950147665 hasConcept C2983860417 @default.
- W2950147665 hasConcept C2984842247 @default.
- W2950147665 hasConcept C33923547 @default.
- W2950147665 hasConcept C37736160 @default.
- W2950147665 hasConcept C38652104 @default.
- W2950147665 hasConcept C41008148 @default.
- W2950147665 hasConcept C55493867 @default.
- W2950147665 hasConcept C63479239 @default.
- W2950147665 hasConcept C6528762 @default.
- W2950147665 hasConcept C68339613 @default.
- W2950147665 hasConcept C70518039 @default.
- W2950147665 hasConceptScore W2950147665C101738243 @default.
- W2950147665 hasConceptScore W2950147665C104317684 @default.
- W2950147665 hasConceptScore W2950147665C108583219 @default.
- W2950147665 hasConceptScore W2950147665C111030470 @default.
- W2950147665 hasConceptScore W2950147665C11413529 @default.
- W2950147665 hasConceptScore W2950147665C119857082 @default.
- W2950147665 hasConceptScore W2950147665C134306372 @default.
- W2950147665 hasConceptScore W2950147665C139676723 @default.
- W2950147665 hasConceptScore W2950147665C154945302 @default.
- W2950147665 hasConceptScore W2950147665C173608175 @default.
- W2950147665 hasConceptScore W2950147665C185592680 @default.
- W2950147665 hasConceptScore W2950147665C2983860417 @default.
- W2950147665 hasConceptScore W2950147665C2984842247 @default.
- W2950147665 hasConceptScore W2950147665C33923547 @default.
- W2950147665 hasConceptScore W2950147665C37736160 @default.
- W2950147665 hasConceptScore W2950147665C38652104 @default.
- W2950147665 hasConceptScore W2950147665C41008148 @default.
- W2950147665 hasConceptScore W2950147665C55493867 @default.
- W2950147665 hasConceptScore W2950147665C63479239 @default.
- W2950147665 hasConceptScore W2950147665C6528762 @default.
- W2950147665 hasConceptScore W2950147665C68339613 @default.
- W2950147665 hasConceptScore W2950147665C70518039 @default.
- W2950147665 hasLocation W29501476651 @default.
- W2950147665 hasOpenAccess W2950147665 @default.
- W2950147665 hasPrimaryLocation W29501476651 @default.
- W2950147665 hasRelatedWork W2773246606 @default.
- W2950147665 hasRelatedWork W2785693366 @default.
- W2950147665 hasRelatedWork W2805329444 @default.
- W2950147665 hasRelatedWork W2985390828 @default.
- W2950147665 hasRelatedWork W2990850284 @default.
- W2950147665 hasRelatedWork W2996490176 @default.
- W2950147665 hasRelatedWork W3006738123 @default.
- W2950147665 hasRelatedWork W3033202076 @default.
- W2950147665 hasRelatedWork W3101202235 @default.
- W2950147665 hasRelatedWork W3106220303 @default.
- W2950147665 hasRelatedWork W3106286647 @default.
- W2950147665 hasRelatedWork W3107524501 @default.
- W2950147665 hasRelatedWork W3117928012 @default.
- W2950147665 hasRelatedWork W3149093307 @default.
- W2950147665 hasRelatedWork W3154935500 @default.
- W2950147665 hasRelatedWork W3155862922 @default.
- W2950147665 hasRelatedWork W3174102408 @default.
- W2950147665 hasRelatedWork W3195826240 @default.
- W2950147665 hasRelatedWork W3198892687 @default.
- W2950147665 hasRelatedWork W3207021344 @default.