Matches in SemOpenAlex for { <https://semopenalex.org/work/W4309802241> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4309802241 abstract "Human Activity Recognition (HAR) has been employed in a wide range of applications, e.g. self-driving cars, where safety and lives are at stake. Recently, the robustness of existing skeleton-based HAR methods has been questioned due to their vulnerability to adversarial attacks, which causes concerns considering the scale of the implication. However, the proposed attacks require the full-knowledge of the attacked classifier, which is overly restrictive. In this paper, we show such threats indeed exist, even when the attacker only has access to the input/output of the model. To this end, we propose the very first black-box adversarial attack approach in skeleton-based HAR called BASAR. BASAR explores the interplay between the classification boundary and the natural motion manifold. To our best knowledge, this is the first time data manifold is introduced in adversarial attacks on time series. Via BASAR, we find on-manifold adversarial samples are extremely deceitful and rather common in skeletal motions, in contrast to the common belief that adversarial samples only exist off-manifold. Through exhaustive evaluation, we show that BASAR can deliver successful attacks across classifiers, datasets, and attack modes. By attack, BASAR helps identify the potential causes of the model vulnerability and provides insights on possible improvements. Finally, to mitigate the newly identified threat, we propose a new adversarial training approach by leveraging the sophisticated distributions of on/off-manifold adversarial samples, called mixed manifold-based adversarial training (MMAT). MMAT can successfully help defend against adversarial attacks without compromising classification accuracy." @default.
- W4309802241 created "2022-11-29" @default.
- W4309802241 creator A5021471823 @default.
- W4309802241 creator A5025226631 @default.
- W4309802241 creator A5026338785 @default.
- W4309802241 creator A5053342436 @default.
- W4309802241 creator A5060069116 @default.
- W4309802241 creator A5060286507 @default.
- W4309802241 date "2022-11-21" @default.
- W4309802241 modified "2023-09-23" @default.
- W4309802241 title "Understanding the Vulnerability of Skeleton-based Human Activity Recognition via Black-box Attack" @default.
- W4309802241 doi "https://doi.org/10.48550/arxiv.2211.11312" @default.
- W4309802241 hasPublicationYear "2022" @default.
- W4309802241 type Work @default.
- W4309802241 citedByCount "0" @default.
- W4309802241 crossrefType "posted-content" @default.
- W4309802241 hasAuthorship W4309802241A5021471823 @default.
- W4309802241 hasAuthorship W4309802241A5025226631 @default.
- W4309802241 hasAuthorship W4309802241A5026338785 @default.
- W4309802241 hasAuthorship W4309802241A5053342436 @default.
- W4309802241 hasAuthorship W4309802241A5060069116 @default.
- W4309802241 hasAuthorship W4309802241A5060286507 @default.
- W4309802241 hasBestOaLocation W43098022411 @default.
- W4309802241 hasConcept C104317684 @default.
- W4309802241 hasConcept C119857082 @default.
- W4309802241 hasConcept C127413603 @default.
- W4309802241 hasConcept C154945302 @default.
- W4309802241 hasConcept C185592680 @default.
- W4309802241 hasConcept C37736160 @default.
- W4309802241 hasConcept C38652104 @default.
- W4309802241 hasConcept C41008148 @default.
- W4309802241 hasConcept C529865628 @default.
- W4309802241 hasConcept C55493867 @default.
- W4309802241 hasConcept C63479239 @default.
- W4309802241 hasConcept C78519656 @default.
- W4309802241 hasConcept C95623464 @default.
- W4309802241 hasConcept C95713431 @default.
- W4309802241 hasConceptScore W4309802241C104317684 @default.
- W4309802241 hasConceptScore W4309802241C119857082 @default.
- W4309802241 hasConceptScore W4309802241C127413603 @default.
- W4309802241 hasConceptScore W4309802241C154945302 @default.
- W4309802241 hasConceptScore W4309802241C185592680 @default.
- W4309802241 hasConceptScore W4309802241C37736160 @default.
- W4309802241 hasConceptScore W4309802241C38652104 @default.
- W4309802241 hasConceptScore W4309802241C41008148 @default.
- W4309802241 hasConceptScore W4309802241C529865628 @default.
- W4309802241 hasConceptScore W4309802241C55493867 @default.
- W4309802241 hasConceptScore W4309802241C63479239 @default.
- W4309802241 hasConceptScore W4309802241C78519656 @default.
- W4309802241 hasConceptScore W4309802241C95623464 @default.
- W4309802241 hasConceptScore W4309802241C95713431 @default.
- W4309802241 hasLocation W43098022411 @default.
- W4309802241 hasOpenAccess W4309802241 @default.
- W4309802241 hasPrimaryLocation W43098022411 @default.
- W4309802241 hasRelatedWork W2556319748 @default.
- W4309802241 hasRelatedWork W3046843850 @default.
- W4309802241 hasRelatedWork W3100995366 @default.
- W4309802241 hasRelatedWork W3200179079 @default.
- W4309802241 hasRelatedWork W3205128835 @default.
- W4309802241 hasRelatedWork W4286899967 @default.
- W4309802241 hasRelatedWork W4293092754 @default.
- W4309802241 hasRelatedWork W4311734044 @default.
- W4309802241 hasRelatedWork W4313887253 @default.
- W4309802241 hasRelatedWork W4379258830 @default.
- W4309802241 isParatext "false" @default.
- W4309802241 isRetracted "false" @default.
- W4309802241 workType "article" @default.