Matches in SemOpenAlex for { <https://semopenalex.org/work/W4289552611> ?p ?o ?g. }
Showing items 1 to 57 of
57
with 100 items per page.
- W4289552611 abstract "In the last decade, deep learning algorithms have become very popular thanks to the achieved performance in many machine learning and computer vision tasks. However, most of the deep learning architectures are vulnerable to so called adversarial examples. This questions the security of deep neural networks (DNN) for many security- and trust-sensitive domains. The majority of the proposed existing adversarial attacks are based on the differentiability of the DNN cost function.Defence strategies are mostly based on machine learning and signal processing principles that either try to detect-reject or filter out the adversarial perturbations and completely neglect the classical cryptographic component in the defence. In this work, we propose a new defence mechanism based on the second Kerckhoffs's cryptographic principle which states that the defence and classification algorithm are supposed to be known, but not the key. To be compliant with the assumption that the attacker does not have access to the secret key, we will primarily focus on a gray-box scenario and do not address a white-box one. More particularly, we assume that the attacker does not have direct access to the secret block, but (a) he completely knows the system architecture, (b) he has access to the data used for training and testing and (c) he can observe the output of the classifier for each given input. We show empirically that our system is efficient against most famous state-of-the-art attacks in black-box and gray-box scenarios." @default.
- W4289552611 created "2022-08-03" @default.
- W4289552611 creator A5008678783 @default.
- W4289552611 creator A5026953476 @default.
- W4289552611 creator A5091506990 @default.
- W4289552611 date "2018-09-05" @default.
- W4289552611 modified "2023-10-05" @default.
- W4289552611 title "Bridging machine learning and cryptography in defence against adversarial attacks" @default.
- W4289552611 doi "https://doi.org/10.48550/arxiv.1809.01715" @default.
- W4289552611 hasPublicationYear "2018" @default.
- W4289552611 type Work @default.
- W4289552611 citedByCount "0" @default.
- W4289552611 crossrefType "posted-content" @default.
- W4289552611 hasAuthorship W4289552611A5008678783 @default.
- W4289552611 hasAuthorship W4289552611A5026953476 @default.
- W4289552611 hasAuthorship W4289552611A5091506990 @default.
- W4289552611 hasBestOaLocation W42895526111 @default.
- W4289552611 hasConcept C108583219 @default.
- W4289552611 hasConcept C119857082 @default.
- W4289552611 hasConcept C147343967 @default.
- W4289552611 hasConcept C154945302 @default.
- W4289552611 hasConcept C174348530 @default.
- W4289552611 hasConcept C178489894 @default.
- W4289552611 hasConcept C2778403875 @default.
- W4289552611 hasConcept C37736160 @default.
- W4289552611 hasConcept C38652104 @default.
- W4289552611 hasConcept C41008148 @default.
- W4289552611 hasConcept C80444323 @default.
- W4289552611 hasConcept C95623464 @default.
- W4289552611 hasConceptScore W4289552611C108583219 @default.
- W4289552611 hasConceptScore W4289552611C119857082 @default.
- W4289552611 hasConceptScore W4289552611C147343967 @default.
- W4289552611 hasConceptScore W4289552611C154945302 @default.
- W4289552611 hasConceptScore W4289552611C174348530 @default.
- W4289552611 hasConceptScore W4289552611C178489894 @default.
- W4289552611 hasConceptScore W4289552611C2778403875 @default.
- W4289552611 hasConceptScore W4289552611C37736160 @default.
- W4289552611 hasConceptScore W4289552611C38652104 @default.
- W4289552611 hasConceptScore W4289552611C41008148 @default.
- W4289552611 hasConceptScore W4289552611C80444323 @default.
- W4289552611 hasConceptScore W4289552611C95623464 @default.
- W4289552611 hasLocation W42895526111 @default.
- W4289552611 hasOpenAccess W4289552611 @default.
- W4289552611 hasPrimaryLocation W42895526111 @default.
- W4289552611 hasRelatedWork W2891797126 @default.
- W4289552611 hasRelatedWork W2913608505 @default.
- W4289552611 hasRelatedWork W2952541330 @default.
- W4289552611 hasRelatedWork W2953083558 @default.
- W4289552611 hasRelatedWork W2963115223 @default.
- W4289552611 hasRelatedWork W3009460750 @default.
- W4289552611 hasRelatedWork W3116352955 @default.
- W4289552611 hasRelatedWork W3205128835 @default.
- W4289552611 hasRelatedWork W4247200422 @default.
- W4289552611 hasRelatedWork W4286899967 @default.
- W4289552611 isParatext "false" @default.
- W4289552611 isRetracted "false" @default.
- W4289552611 workType "article" @default.