Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313484756> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W4313484756 abstract "Because of their high accuracy, deep neural net-works (DNNs) have achieved amazing success in security-critical systems such as medical devices. It has recently been demon-strated that Adversarial Bit Flip Attacks (BFAs) against DNN hardware by flipping a very small number of bits can result in catastrophic accuracy loss. The reliance on test data, however, is a significant drawback of previous state-of-the-art bit-flip attack methods. This is frequently not possible with applications containing sensitive or proprietary data. In this paper, we propose Blind Data Adversarial Bit-flip Attack (BDFA), a novel technique to enable BFA against DNN hardware without any access to the training or testing data. This is achieved by optimizing for a synthetic dataset, which is engineered to match the statistics of batch normalization across different layers of the network and the targeted label. Experimental results show that BDFA could decrease the accuracy of ResNet50 significantly from 75.96% to 13.94% with only 4 bits flips." @default.
- W4313484756 created "2023-01-06" @default.
- W4313484756 creator A5027508220 @default.
- W4313484756 creator A5028010546 @default.
- W4313484756 creator A5062627025 @default.
- W4313484756 creator A5065889904 @default.
- W4313484756 creator A5083398015 @default.
- W4313484756 date "2022-08-01" @default.
- W4313484756 modified "2023-09-30" @default.
- W4313484756 title "Blind Data Adversarial Bit-flip Attack against Deep Neural Networks" @default.
- W4313484756 cites W2157116240 @default.
- W4313484756 cites W2243397390 @default.
- W4313484756 cites W2789304371 @default.
- W4313484756 cites W2807835252 @default.
- W4313484756 cites W2963165363 @default.
- W4313484756 cites W2963857521 @default.
- W4313484756 cites W2963946985 @default.
- W4313484756 cites W2981860227 @default.
- W4313484756 cites W3034940165 @default.
- W4313484756 cites W3046853140 @default.
- W4313484756 cites W3132455321 @default.
- W4313484756 cites W3139156009 @default.
- W4313484756 cites W3159196909 @default.
- W4313484756 cites W4242053016 @default.
- W4313484756 cites W4288337628 @default.
- W4313484756 doi "https://doi.org/10.1109/dsd57027.2022.00126" @default.
- W4313484756 hasPublicationYear "2022" @default.
- W4313484756 type Work @default.
- W4313484756 citedByCount "2" @default.
- W4313484756 countsByYear W43134847562023 @default.
- W4313484756 crossrefType "proceedings-article" @default.
- W4313484756 hasAuthorship W4313484756A5027508220 @default.
- W4313484756 hasAuthorship W4313484756A5028010546 @default.
- W4313484756 hasAuthorship W4313484756A5062627025 @default.
- W4313484756 hasAuthorship W4313484756A5065889904 @default.
- W4313484756 hasAuthorship W4313484756A5083398015 @default.
- W4313484756 hasConcept C113775141 @default.
- W4313484756 hasConcept C11413529 @default.
- W4313484756 hasConcept C117011727 @default.
- W4313484756 hasConcept C136886441 @default.
- W4313484756 hasConcept C144024400 @default.
- W4313484756 hasConcept C153180895 @default.
- W4313484756 hasConcept C154945302 @default.
- W4313484756 hasConcept C19165224 @default.
- W4313484756 hasConcept C2984842247 @default.
- W4313484756 hasConcept C31258907 @default.
- W4313484756 hasConcept C37736160 @default.
- W4313484756 hasConcept C41008148 @default.
- W4313484756 hasConcept C50644808 @default.
- W4313484756 hasConceptScore W4313484756C113775141 @default.
- W4313484756 hasConceptScore W4313484756C11413529 @default.
- W4313484756 hasConceptScore W4313484756C117011727 @default.
- W4313484756 hasConceptScore W4313484756C136886441 @default.
- W4313484756 hasConceptScore W4313484756C144024400 @default.
- W4313484756 hasConceptScore W4313484756C153180895 @default.
- W4313484756 hasConceptScore W4313484756C154945302 @default.
- W4313484756 hasConceptScore W4313484756C19165224 @default.
- W4313484756 hasConceptScore W4313484756C2984842247 @default.
- W4313484756 hasConceptScore W4313484756C31258907 @default.
- W4313484756 hasConceptScore W4313484756C37736160 @default.
- W4313484756 hasConceptScore W4313484756C41008148 @default.
- W4313484756 hasConceptScore W4313484756C50644808 @default.
- W4313484756 hasLocation W43134847561 @default.
- W4313484756 hasOpenAccess W4313484756 @default.
- W4313484756 hasPrimaryLocation W43134847561 @default.
- W4313484756 hasRelatedWork W1991269640 @default.
- W4313484756 hasRelatedWork W2016839265 @default.
- W4313484756 hasRelatedWork W2516800609 @default.
- W4313484756 hasRelatedWork W2942801205 @default.
- W4313484756 hasRelatedWork W2995319142 @default.
- W4313484756 hasRelatedWork W3129887816 @default.
- W4313484756 hasRelatedWork W3200578280 @default.
- W4313484756 hasRelatedWork W4200633666 @default.
- W4313484756 hasRelatedWork W4288359223 @default.
- W4313484756 hasRelatedWork W2508457823 @default.
- W4313484756 isParatext "false" @default.
- W4313484756 isRetracted "false" @default.
- W4313484756 workType "article" @default.