Matches in SemOpenAlex for { <https://semopenalex.org/work/W2787446543> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W2787446543 abstract "Recent advances in adversarial Deep Learning (DL) have opened up a new and largely unexplored surface for malicious attacks jeopardizing the integrity of autonomous DL systems. We introduce a novel automated countermeasure called Parallel Checkpointing Learners (PCL) to thwart the potential adversarial attacks and significantly improve the reliability (safety) of a victim DL model. The proposed PCL methodology is unsupervised, meaning that no adversarial sample is leveraged to build/train parallel checkpointing learners. We formalize the goal of preventing adversarial attacks as an optimization problem to minimize the rarely observed regions in the latent feature space spanned by a DL network. To solve the aforementioned minimization problem, a set of complementary but disjoint checkpointing modules are trained and leveraged to validate the victim model execution in parallel. Each checkpointing learner explicitly characterizes the geometry of the input data and the corresponding high-level data abstractions within a particular DL layer. As such, the adversary is required to simultaneously deceive all the defender modules in order to succeed. We extensively evaluate the performance of the PCL methodology against the state-of-the-art attack scenarios, including Fast-Gradient-Sign (FGS), Jacobian Saliency Map Attack (JSMA), Deepfool, and Carlini&WagnerL2 algorithm. Extensive proof-of-concept evaluations for analyzing various data collections including MNIST, CIFAR10, and ImageNet corroborate the effectiveness of our proposed defense mechanism against adversarial samples." @default.
- W2787446543 created "2018-02-23" @default.
- W2787446543 creator A5033081533 @default.
- W2787446543 creator A5035073741 @default.
- W2787446543 creator A5059310658 @default.
- W2787446543 creator A5063273639 @default.
- W2787446543 date "2018-02-15" @default.
- W2787446543 modified "2023-09-23" @default.
- W2787446543 title "Towards Safe Deep Learning: Unsupervised Defense Against Generic Adversarial Attacks" @default.
- W2787446543 cites W1883420340 @default.
- W2787446543 cites W1966948031 @default.
- W2787446543 cites W2108598243 @default.
- W2787446543 cites W2112796928 @default.
- W2787446543 cites W2127271355 @default.
- W2787446543 cites W2243397390 @default.
- W2787446543 cites W2963207607 @default.
- W2787446543 cites W2963857521 @default.
- W2787446543 cites W2964040467 @default.
- W2787446543 cites W2964153729 @default.
- W2787446543 cites W3118608800 @default.
- W2787446543 hasPublicationYear "2018" @default.
- W2787446543 type Work @default.
- W2787446543 sameAs 2787446543 @default.
- W2787446543 citedByCount "2" @default.
- W2787446543 countsByYear W27874465432019 @default.
- W2787446543 crossrefType "journal-article" @default.
- W2787446543 hasAuthorship W2787446543A5033081533 @default.
- W2787446543 hasAuthorship W2787446543A5035073741 @default.
- W2787446543 hasAuthorship W2787446543A5059310658 @default.
- W2787446543 hasAuthorship W2787446543A5063273639 @default.
- W2787446543 hasConcept C108583219 @default.
- W2787446543 hasConcept C114614502 @default.
- W2787446543 hasConcept C119857082 @default.
- W2787446543 hasConcept C154945302 @default.
- W2787446543 hasConcept C177264268 @default.
- W2787446543 hasConcept C190502265 @default.
- W2787446543 hasConcept C199360897 @default.
- W2787446543 hasConcept C33923547 @default.
- W2787446543 hasConcept C37736160 @default.
- W2787446543 hasConcept C38652104 @default.
- W2787446543 hasConcept C41008148 @default.
- W2787446543 hasConcept C41065033 @default.
- W2787446543 hasConcept C45340560 @default.
- W2787446543 hasConceptScore W2787446543C108583219 @default.
- W2787446543 hasConceptScore W2787446543C114614502 @default.
- W2787446543 hasConceptScore W2787446543C119857082 @default.
- W2787446543 hasConceptScore W2787446543C154945302 @default.
- W2787446543 hasConceptScore W2787446543C177264268 @default.
- W2787446543 hasConceptScore W2787446543C190502265 @default.
- W2787446543 hasConceptScore W2787446543C199360897 @default.
- W2787446543 hasConceptScore W2787446543C33923547 @default.
- W2787446543 hasConceptScore W2787446543C37736160 @default.
- W2787446543 hasConceptScore W2787446543C38652104 @default.
- W2787446543 hasConceptScore W2787446543C41008148 @default.
- W2787446543 hasConceptScore W2787446543C41065033 @default.
- W2787446543 hasConceptScore W2787446543C45340560 @default.
- W2787446543 hasOpenAccess W2787446543 @default.
- W2787446543 hasRelatedWork W2617106563 @default.
- W2787446543 hasRelatedWork W2773246606 @default.
- W2787446543 hasRelatedWork W2969333443 @default.
- W2787446543 hasRelatedWork W2990850284 @default.
- W2787446543 hasRelatedWork W2996490176 @default.
- W2787446543 hasRelatedWork W3004923651 @default.
- W2787446543 hasRelatedWork W3023057171 @default.
- W2787446543 hasRelatedWork W3034325757 @default.
- W2787446543 hasRelatedWork W3038342492 @default.
- W2787446543 hasRelatedWork W3086943318 @default.
- W2787446543 hasRelatedWork W3088237946 @default.
- W2787446543 hasRelatedWork W3101882350 @default.
- W2787446543 hasRelatedWork W3102631191 @default.
- W2787446543 hasRelatedWork W3106286647 @default.
- W2787446543 hasRelatedWork W3114686421 @default.
- W2787446543 hasRelatedWork W3117928012 @default.
- W2787446543 hasRelatedWork W3195826240 @default.
- W2787446543 hasRelatedWork W3207021344 @default.
- W2787446543 hasRelatedWork W3210016964 @default.
- W2787446543 hasRelatedWork W3212298887 @default.
- W2787446543 isParatext "false" @default.
- W2787446543 isRetracted "false" @default.
- W2787446543 magId "2787446543" @default.
- W2787446543 workType "article" @default.