FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2964201752> ?p ?o ?g. }

• W2964201752 abstract "We introduce a feature scattering-based adversarial training approach for improving model robustness against adversarial attacks. Conventional adversarial training approaches leverage a supervised scheme (either targeted or non-targeted) in generating attacks for training, which typically suffer from issues such as label leaking as noted in recent works. Differently, the proposed approach generates adversarial images for training through feature scattering in the latent space, which is unsupervised in nature and avoids label leaking. More importantly, this new approach generates perturbed images in a collaborative fashion, taking the inter-sample relationships into consideration. We conduct analysis on model robustness and demonstrate the effectiveness of the proposed approach through extensively experiments on different datasets compared with state-of-the-art approaches." @default.
• W2964201752 created "2019-07-30" @default.
• W2964201752 creator A5050613147 @default.
• W2964201752 creator A5052624521 @default.
• W2964201752 date "2019-07-24" @default.
• W2964201752 modified "2023-09-27" @default.
• W2964201752 title "Defense Against Adversarial Attacks Using Feature Scattering-based Adversarial Training" @default.
• W2964201752 cites W1676470423 @default.
• W2964201752 cites W205159212 @default.
• W2964201752 cites W2063532964 @default.
• W2964201752 cites W2099471712 @default.
• W2964201752 cites W2100495367 @default.
• W2964201752 cites W2144935315 @default.
• W2964201752 cites W2158131535 @default.
• W2964201752 cites W2243397390 @default.
• W2964201752 cites W2335728318 @default.
• W2964201752 cites W2416516526 @default.
• W2964201752 cites W2417426974 @default.
• W2964201752 cites W2513314332 @default.
• W2964201752 cites W2543927648 @default.
• W2964201752 cites W2593892853 @default.
• W2964201752 cites W2604505099 @default.
• W2964201752 cites W2606711863 @default.
• W2964201752 cites W2612637113 @default.
• W2964201752 cites W2618043096 @default.
• W2964201752 cites W2623516925 @default.
• W2964201752 cites W2738841453 @default.
• W2964201752 cites W2739748921 @default.
• W2964201752 cites W2765233338 @default.
• W2964201752 cites W2773446523 @default.
• W2964201752 cites W2774018344 @default.
• W2964201752 cites W2778624544 @default.
• W2964201752 cites W2783555701 @default.
• W2964201752 cites W2786002645 @default.
• W2964201752 cites W2791953061 @default.
• W2964201752 cites W2793477525 @default.
• W2964201752 cites W2794002979 @default.
• W2964201752 cites W2799032899 @default.
• W2964201752 cites W2799152959 @default.
• W2964201752 cites W2884519271 @default.
• W2964201752 cites W2891641674 @default.
• W2964201752 cites W2903785932 @default.
• W2964201752 cites W2905418437 @default.
• W2964201752 cites W2909303050 @default.
• W2964201752 cites W2920241157 @default.
• W2964201752 cites W2942630857 @default.
• W2964201752 cites W2944621111 @default.
• W2964201752 cites W2945033152 @default.
• W2964201752 cites W2949152835 @default.
• W2964201752 cites W2962710014 @default.
• W2964201752 cites W2962768284 @default.
• W2964201752 cites W2962933288 @default.
• W2964201752 cites W2963025848 @default.
• W2964201752 cites W2963070423 @default.
• W2964201752 cites W2963143631 @default.
• W2964201752 cites W2963207607 @default.
• W2964201752 cites W2963389226 @default.
• W2964201752 cites W2963399829 @default.
• W2964201752 cites W2963506485 @default.
• W2964201752 cites W2963566318 @default.
• W2964201752 cites W2963744840 @default.
• W2964201752 cites W2963746531 @default.
• W2964201752 cites W2963857521 @default.
• W2964201752 cites W2963915682 @default.
• W2964201752 cites W2963920068 @default.
• W2964201752 cites W2963989027 @default.
• W2964201752 cites W2964137095 @default.
• W2964201752 cites W2964153729 @default.
• W2964201752 cites W2964197269 @default.
• W2964201752 cites W2964253222 @default.
• W2964201752 cites W2964301649 @default.
• W2964201752 cites W2981495453 @default.
• W2964201752 cites W3004298045 @default.
• W2964201752 cites W3103557498 @default.
• W2964201752 cites W3118608800 @default.
• W2964201752 cites W385466589 @default.
• W2964201752 cites W9657784 @default.
• W2964201752 hasPublicationYear "2019" @default.
• W2964201752 type Work @default.
• W2964201752 sameAs 2964201752 @default.
• W2964201752 citedByCount "9" @default.
• W2964201752 countsByYear W29642017522018 @default.
• W2964201752 countsByYear W29642017522019 @default.
• W2964201752 countsByYear W29642017522020 @default.
• W2964201752 countsByYear W29642017522021 @default.
• W2964201752 crossrefType "posted-content" @default.
• W2964201752 hasAuthorship W2964201752A5050613147 @default.
• W2964201752 hasAuthorship W2964201752A5052624521 @default.
• W2964201752 hasConcept C104317684 @default.
• W2964201752 hasConcept C119857082 @default.
• W2964201752 hasConcept C138885662 @default.
• W2964201752 hasConcept C153083717 @default.
• W2964201752 hasConcept C153180895 @default.
• W2964201752 hasConcept C154945302 @default.
• W2964201752 hasConcept C185592680 @default.
• W2964201752 hasConcept C2776401178 @default.
• W2964201752 hasConcept C37736160 @default.
• W2964201752 hasConcept C38652104 @default.
• W2964201752 hasConcept C41008148 @default.
• W2964201752 hasConcept C41065033 @default.

• next