FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4383346107> ?p ?o ?g. }

• W4383346107 endingPage "6173" @default.
• W4383346107 startingPage "6173" @default.
• W4383346107 abstract "Adversarial attacks have become one of the most serious security issues in widely used deep neural networks. Even though real-world datasets usually have large intra-variations or multiple modes, most adversarial defense methods, such as adversarial training, which is currently one of the most effective defense methods, mainly focus on the single-mode setting and thus fail to capture the full data representation to defend against adversarial attacks. To confront this challenge, we propose a novel multi-prototype metric learning regularization for adversarial training which can effectively enhance the defense capability of adversarial training by preventing the latent representation of the adversarial example changing a lot from its clean one. With extensive experiments on CIFAR10, CIFAR100, MNIST, and Tiny ImageNet, the evaluation results show the proposed method improves the performance of different state-of-the-art adversarial training methods without additional computational cost. Furthermore, besides Tiny ImageNet, in the multi-prototype CIFAR10 and CIFAR100 where we reorganize the whole datasets of CIFAR10 and CIFAR100 into two and ten classes, respectively, the proposed method outperforms the state-of-the-art approach by 2.22% and 1.65%, respectively. Furthermore, the proposed multi-prototype method also outperforms its single-prototype version and other commonly used deep metric learning approaches as regularization for adversarial training and thus further demonstrates its effectiveness." @default.
• W4383346107 created "2023-07-07" @default.
• W4383346107 creator A5005486605 @default.
• W4383346107 creator A5010007692 @default.
• W4383346107 creator A5072890368 @default.
• W4383346107 creator A5084278510 @default.
• W4383346107 date "2023-07-05" @default.
• W4383346107 modified "2023-09-30" @default.
• W4383346107 title "Towards Adversarial Robustness for Multi-Mode Data through Metric Learning" @default.
• W4383346107 cites W1932198206 @default.
• W4383346107 cites W2097117768 @default.
• W4383346107 cites W2112796928 @default.
• W4383346107 cites W2180612164 @default.
• W4383346107 cites W2194775991 @default.
• W4383346107 cites W2605102252 @default.
• W4383346107 cites W2774644650 @default.
• W4383346107 cites W2798303923 @default.
• W4383346107 cites W2919115771 @default.
• W4383346107 cites W2962872506 @default.
• W4383346107 cites W2963026686 @default.
• W4383346107 cites W2963775347 @default.
• W4383346107 cites W2963857521 @default.
• W4383346107 cites W2963988212 @default.
• W4383346107 cites W2964082701 @default.
• W4383346107 cites W2964137095 @default.
• W4383346107 cites W2966048283 @default.
• W4383346107 cites W2969656782 @default.
• W4383346107 cites W3034202663 @default.
• W4383346107 cites W3034303554 @default.
• W4383346107 cites W3106250896 @default.
• W4383346107 cites W3172507542 @default.
• W4383346107 cites W3180134609 @default.
• W4383346107 cites W4251147723 @default.
• W4383346107 cites W4319300873 @default.
• W4383346107 cites W4323925641 @default.
• W4383346107 cites W4377081716 @default.
• W4383346107 doi "https://doi.org/10.3390/s23136173" @default.
• W4383346107 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37448021" @default.
• W4383346107 hasPublicationYear "2023" @default.
• W4383346107 type Work @default.
• W4383346107 citedByCount "0" @default.
• W4383346107 crossrefType "journal-article" @default.
• W4383346107 hasAuthorship W4383346107A5005486605 @default.
• W4383346107 hasAuthorship W4383346107A5010007692 @default.
• W4383346107 hasAuthorship W4383346107A5072890368 @default.
• W4383346107 hasAuthorship W4383346107A5084278510 @default.
• W4383346107 hasBestOaLocation W43833461071 @default.
• W4383346107 hasConcept C104317684 @default.
• W4383346107 hasConcept C108583219 @default.
• W4383346107 hasConcept C119857082 @default.
• W4383346107 hasConcept C127413603 @default.
• W4383346107 hasConcept C154945302 @default.
• W4383346107 hasConcept C176217482 @default.
• W4383346107 hasConcept C17744445 @default.
• W4383346107 hasConcept C185592680 @default.
• W4383346107 hasConcept C190502265 @default.
• W4383346107 hasConcept C199539241 @default.
• W4383346107 hasConcept C21547014 @default.
• W4383346107 hasConcept C2776135515 @default.
• W4383346107 hasConcept C2776359362 @default.
• W4383346107 hasConcept C2984842247 @default.
• W4383346107 hasConcept C37736160 @default.
• W4383346107 hasConcept C41008148 @default.
• W4383346107 hasConcept C55493867 @default.
• W4383346107 hasConcept C63479239 @default.
• W4383346107 hasConcept C94625758 @default.
• W4383346107 hasConceptScore W4383346107C104317684 @default.
• W4383346107 hasConceptScore W4383346107C108583219 @default.
• W4383346107 hasConceptScore W4383346107C119857082 @default.
• W4383346107 hasConceptScore W4383346107C127413603 @default.
• W4383346107 hasConceptScore W4383346107C154945302 @default.
• W4383346107 hasConceptScore W4383346107C176217482 @default.
• W4383346107 hasConceptScore W4383346107C17744445 @default.
• W4383346107 hasConceptScore W4383346107C185592680 @default.
• W4383346107 hasConceptScore W4383346107C190502265 @default.
• W4383346107 hasConceptScore W4383346107C199539241 @default.
• W4383346107 hasConceptScore W4383346107C21547014 @default.
• W4383346107 hasConceptScore W4383346107C2776135515 @default.
• W4383346107 hasConceptScore W4383346107C2776359362 @default.
• W4383346107 hasConceptScore W4383346107C2984842247 @default.
• W4383346107 hasConceptScore W4383346107C37736160 @default.
• W4383346107 hasConceptScore W4383346107C41008148 @default.
• W4383346107 hasConceptScore W4383346107C55493867 @default.
• W4383346107 hasConceptScore W4383346107C63479239 @default.
• W4383346107 hasConceptScore W4383346107C94625758 @default.
• W4383346107 hasIssue "13" @default.
• W4383346107 hasLocation W43833461071 @default.
• W4383346107 hasLocation W43833461072 @default.
• W4383346107 hasLocation W43833461073 @default.
• W4383346107 hasOpenAccess W4383346107 @default.
• W4383346107 hasPrimaryLocation W43833461071 @default.
• W4383346107 hasRelatedWork W2798548091 @default.
• W4383346107 hasRelatedWork W2967955854 @default.
• W4383346107 hasRelatedWork W2973831015 @default.
• W4383346107 hasRelatedWork W2989784533 @default.
• W4383346107 hasRelatedWork W3127679336 @default.
• W4383346107 hasRelatedWork W3193857078 @default.
• W4383346107 hasRelatedWork W3208723233 @default.

• next