Matches in SemOpenAlex for { <https://semopenalex.org/work/W3105904398> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W3105904398 abstract "Adversarial examples have emerged as a significant threat to machine learning algorithms, especially to the convolutional neural networks (CNNs). In this paper, we propose two quantization-based defense mechanisms, Constant Quantization (CQ) and Trainable Quantization (TQ), to increase the robustness of CNNs against adversarial examples. CQ quantizes input pixel intensities based on a fixed number of quantization levels, while in TQ, the quantization levels are iteratively learned during the training phase, thereby providing a stronger defense mechanism. We apply the proposed techniques on undefended CNNs against different state-of-the-art adversarial attacks from the open-source textit{Cleverhans} library. The experimental results demonstrate 50%-96% and 10%-50% increase in the classification accuracy of the perturbed images generated from the MNIST and the CIFAR-10 datasets, respectively, on commonly used CNN (Conv2D(64, 8x8) - Conv2D(128, 6x6) - Conv2D(128, 5x5) - Dense(10) - Softmax()) available in textit{Cleverhans} library." @default.
- W3105904398 created "2020-11-23" @default.
- W3105904398 creator A5005190949 @default.
- W3105904398 creator A5023695007 @default.
- W3105904398 creator A5047007749 @default.
- W3105904398 creator A5058712739 @default.
- W3105904398 creator A5066402727 @default.
- W3105904398 creator A5074265316 @default.
- W3105904398 creator A5087677488 @default.
- W3105904398 date "2019-07-01" @default.
- W3105904398 modified "2023-10-18" @default.
- W3105904398 title "QuSecNets: Quantization-based Defense Mechanism for Securing Deep Neural Network against Adversarial Attacks" @default.
- W3105904398 cites W2243397390 @default.
- W3105904398 cites W2607219512 @default.
- W3105904398 cites W2607757716 @default.
- W3105904398 cites W2618043096 @default.
- W3105904398 cites W2798919674 @default.
- W3105904398 cites W2798993323 @default.
- W3105904398 cites W2801550278 @default.
- W3105904398 cites W2886576643 @default.
- W3105904398 cites W2897317743 @default.
- W3105904398 cites W2898009658 @default.
- W3105904398 cites W2940718416 @default.
- W3105904398 cites W2963178695 @default.
- W3105904398 cites W2963626858 @default.
- W3105904398 cites W2964082701 @default.
- W3105904398 doi "https://doi.org/10.1109/iolts.2019.8854377" @default.
- W3105904398 hasPublicationYear "2019" @default.
- W3105904398 type Work @default.
- W3105904398 sameAs 3105904398 @default.
- W3105904398 citedByCount "21" @default.
- W3105904398 countsByYear W31059043982020 @default.
- W3105904398 countsByYear W31059043982021 @default.
- W3105904398 countsByYear W31059043982022 @default.
- W3105904398 countsByYear W31059043982023 @default.
- W3105904398 crossrefType "proceedings-article" @default.
- W3105904398 hasAuthorship W3105904398A5005190949 @default.
- W3105904398 hasAuthorship W3105904398A5023695007 @default.
- W3105904398 hasAuthorship W3105904398A5047007749 @default.
- W3105904398 hasAuthorship W3105904398A5058712739 @default.
- W3105904398 hasAuthorship W3105904398A5066402727 @default.
- W3105904398 hasAuthorship W3105904398A5074265316 @default.
- W3105904398 hasAuthorship W3105904398A5087677488 @default.
- W3105904398 hasBestOaLocation W31059043982 @default.
- W3105904398 hasConcept C104317684 @default.
- W3105904398 hasConcept C11413529 @default.
- W3105904398 hasConcept C153180895 @default.
- W3105904398 hasConcept C154945302 @default.
- W3105904398 hasConcept C160633673 @default.
- W3105904398 hasConcept C185592680 @default.
- W3105904398 hasConcept C188441871 @default.
- W3105904398 hasConcept C190502265 @default.
- W3105904398 hasConcept C28855332 @default.
- W3105904398 hasConcept C2984842247 @default.
- W3105904398 hasConcept C37736160 @default.
- W3105904398 hasConcept C41008148 @default.
- W3105904398 hasConcept C50644808 @default.
- W3105904398 hasConcept C55493867 @default.
- W3105904398 hasConcept C63479239 @default.
- W3105904398 hasConcept C81363708 @default.
- W3105904398 hasConceptScore W3105904398C104317684 @default.
- W3105904398 hasConceptScore W3105904398C11413529 @default.
- W3105904398 hasConceptScore W3105904398C153180895 @default.
- W3105904398 hasConceptScore W3105904398C154945302 @default.
- W3105904398 hasConceptScore W3105904398C160633673 @default.
- W3105904398 hasConceptScore W3105904398C185592680 @default.
- W3105904398 hasConceptScore W3105904398C188441871 @default.
- W3105904398 hasConceptScore W3105904398C190502265 @default.
- W3105904398 hasConceptScore W3105904398C28855332 @default.
- W3105904398 hasConceptScore W3105904398C2984842247 @default.
- W3105904398 hasConceptScore W3105904398C37736160 @default.
- W3105904398 hasConceptScore W3105904398C41008148 @default.
- W3105904398 hasConceptScore W3105904398C50644808 @default.
- W3105904398 hasConceptScore W3105904398C55493867 @default.
- W3105904398 hasConceptScore W3105904398C63479239 @default.
- W3105904398 hasConceptScore W3105904398C81363708 @default.
- W3105904398 hasLocation W31059043981 @default.
- W3105904398 hasLocation W31059043982 @default.
- W3105904398 hasOpenAccess W3105904398 @default.
- W3105904398 hasPrimaryLocation W31059043981 @default.
- W3105904398 hasRelatedWork W2798548091 @default.
- W3105904398 hasRelatedWork W2967955854 @default.
- W3105904398 hasRelatedWork W2973831015 @default.
- W3105904398 hasRelatedWork W2989784533 @default.
- W3105904398 hasRelatedWork W3105904398 @default.
- W3105904398 hasRelatedWork W3127679336 @default.
- W3105904398 hasRelatedWork W4288018014 @default.
- W3105904398 hasRelatedWork W4288264694 @default.
- W3105904398 hasRelatedWork W4300326282 @default.
- W3105904398 hasRelatedWork W4308702586 @default.
- W3105904398 isParatext "false" @default.
- W3105904398 isRetracted "false" @default.
- W3105904398 magId "3105904398" @default.
- W3105904398 workType "article" @default.