FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3217492763> ?p ?o ?g. }

• W3217492763 abstract "Deep neural network-based image classification can be misled by adversarial examples with small and quasi-imperceptible perturbations. Furthermore, the adversarial examples created on one classification model can also fool another different model. The transferability of the adversarial examples has recently attracted a growing interest since it makes black-box attacks on classification models feasible. As an extension of classification, semantic segmentation has also received much attention towards its adversarial robustness. However, the transferability of adversarial examples on segmentation models has not been systematically studied. In this work, we intensively study this topic. First, we explore the overfitting phenomenon of adversarial examples on classification and segmentation models. In contrast to the observation made on classification models that the transferability is limited by overfitting to the source model, we find that the adversarial examples on segmentations do not always overfit the source models. Even when no overfitting is presented, the transferability of adversarial examples is limited. We attribute the limitation to the architectural traits of segmentation models, i.e., multi-scale object recognition. Then, we propose a simple and effective method, dubbed dynamic scaling, to overcome the limitation. The high transferability achieved by our method shows that, in contrast to the observations in previous work, adversarial examples on a segmentation model can be easy to transfer to other segmentation models. Our analysis and proposals are supported by extensive experiments." @default.
• W3217492763 created "2021-12-06" @default.
• W3217492763 creator A5042899882 @default.
• W3217492763 creator A5055994909 @default.
• W3217492763 creator A5074808403 @default.
• W3217492763 creator A5078109015 @default.
• W3217492763 date "2021-11-22" @default.
• W3217492763 modified "2023-09-27" @default.
• W3217492763 title "Adversarial Examples on Segmentation Models Can be Easy to Transfer" @default.
• W3217492763 cites W1903029394 @default.
• W3217492763 cites W2031489346 @default.
• W3217492763 cites W2097117768 @default.
• W3217492763 cites W2108598243 @default.
• W3217492763 cites W2194775991 @default.
• W3217492763 cites W2340897893 @default.
• W3217492763 cites W2412782625 @default.
• W3217492763 cites W2560023338 @default.
• W3217492763 cites W2604505099 @default.
• W3217492763 cites W2612445135 @default.
• W3217492763 cites W2630837129 @default.
• W3217492763 cites W2774644650 @default.
• W3217492763 cites W2888706593 @default.
• W3217492763 cites W2890430415 @default.
• W3217492763 cites W2895595267 @default.
• W3217492763 cites W2905573335 @default.
• W3217492763 cites W2962686123 @default.
• W3217492763 cites W2962835968 @default.
• W3217492763 cites W2962847335 @default.
• W3217492763 cites W2962914239 @default.
• W3217492763 cites W2962968216 @default.
• W3217492763 cites W2963118571 @default.
• W3217492763 cites W2963207607 @default.
• W3217492763 cites W2963542245 @default.
• W3217492763 cites W2963840672 @default.
• W3217492763 cites W2963918968 @default.
• W3217492763 cites W2964097310 @default.
• W3217492763 cites W2964153729 @default.
• W3217492763 cites W2964253222 @default.
• W3217492763 cites W2969542116 @default.
• W3217492763 cites W2969945254 @default.
• W3217492763 cites W2970971581 @default.
• W3217492763 cites W2984699060 @default.
• W3217492763 cites W2996140774 @default.
• W3217492763 cites W2997583194 @default.
• W3217492763 cites W3006076803 @default.
• W3217492763 cites W3099873944 @default.
• W3217492763 cites W3102103184 @default.
• W3217492763 cites W3106050153 @default.
• W3217492763 cites W3137894200 @default.
• W3217492763 cites W3167931515 @default.
• W3217492763 cites W3177184533 @default.
• W3217492763 cites W3196107314 @default.
• W3217492763 cites W3203064887 @default.
• W3217492763 cites W9657784 @default.
• W3217492763 doi "https://doi.org/10.48550/arxiv.2111.11368" @default.
• W3217492763 hasPublicationYear "2021" @default.
• W3217492763 type Work @default.
• W3217492763 sameAs 3217492763 @default.
• W3217492763 citedByCount "0" @default.
• W3217492763 crossrefType "posted-content" @default.
• W3217492763 hasAuthorship W3217492763A5042899882 @default.
• W3217492763 hasAuthorship W3217492763A5055994909 @default.
• W3217492763 hasAuthorship W3217492763A5074808403 @default.
• W3217492763 hasAuthorship W3217492763A5078109015 @default.
• W3217492763 hasBestOaLocation W32174927631 @default.
• W3217492763 hasConcept C104317684 @default.
• W3217492763 hasConcept C108583219 @default.
• W3217492763 hasConcept C115961682 @default.
• W3217492763 hasConcept C119857082 @default.
• W3217492763 hasConcept C140331021 @default.
• W3217492763 hasConcept C153180895 @default.
• W3217492763 hasConcept C154945302 @default.
• W3217492763 hasConcept C185592680 @default.
• W3217492763 hasConcept C22019652 @default.
• W3217492763 hasConcept C37736160 @default.
• W3217492763 hasConcept C41008148 @default.
• W3217492763 hasConcept C50644808 @default.
• W3217492763 hasConcept C55493867 @default.
• W3217492763 hasConcept C61272859 @default.
• W3217492763 hasConcept C63479239 @default.
• W3217492763 hasConcept C75294576 @default.
• W3217492763 hasConcept C89600930 @default.
• W3217492763 hasConceptScore W3217492763C104317684 @default.
• W3217492763 hasConceptScore W3217492763C108583219 @default.
• W3217492763 hasConceptScore W3217492763C115961682 @default.
• W3217492763 hasConceptScore W3217492763C119857082 @default.
• W3217492763 hasConceptScore W3217492763C140331021 @default.
• W3217492763 hasConceptScore W3217492763C153180895 @default.
• W3217492763 hasConceptScore W3217492763C154945302 @default.
• W3217492763 hasConceptScore W3217492763C185592680 @default.
• W3217492763 hasConceptScore W3217492763C22019652 @default.
• W3217492763 hasConceptScore W3217492763C37736160 @default.
• W3217492763 hasConceptScore W3217492763C41008148 @default.
• W3217492763 hasConceptScore W3217492763C50644808 @default.
• W3217492763 hasConceptScore W3217492763C55493867 @default.
• W3217492763 hasConceptScore W3217492763C61272859 @default.
• W3217492763 hasConceptScore W3217492763C63479239 @default.
• W3217492763 hasConceptScore W3217492763C75294576 @default.
• W3217492763 hasConceptScore W3217492763C89600930 @default.
• W3217492763 hasLocation W32174927631 @default.

• next