Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312759853> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W4312759853 abstract "In this paper, we study the robustness of graph convolutional networks (GCNs). Previous works have shown that GCNs are vulnerable to adversarial perturbation on adjacency or feature matrices of existing nodes; however, such attacks are usually unrealistic in real applications. For instance, in social network applications, the attacker will need to hack into either the client or server to change existing links or features. In this paper, we propose a new type of “fake node attacks” to attack GCNs by adding malicious fake nodes. This is much more realistic than previous attacks; in social network applications, the attacker only needs to register a set of fake accounts and link to existing ones. To conduct fake node attacks, a greedy algorithm is proposed to generate edges of malicious nodes and their corresponding features aiming to minimize the classification accuracy on the target nodes. In addition, we introduce a discriminator to classify malicious nodes from real nodes and propose a Greedy-generative adversarial network attack to simultaneously update the discriminator and the attacker, to make malicious nodes indistinguishable from the real ones. Our non-targeted attack decreases the accuracy of GCN down to 0.03, and our targeted attack reaches a success rate of 78% on a group of 100 nodes and 90% on average for attacking a single target node." @default.
- W4312759853 created "2023-01-05" @default.
- W4312759853 creator A5000534132 @default.
- W4312759853 creator A5010841999 @default.
- W4312759853 creator A5046791174 @default.
- W4312759853 creator A5073670556 @default.
- W4312759853 creator A5088633029 @default.
- W4312759853 date "2022-10-28" @default.
- W4312759853 modified "2023-10-06" @default.
- W4312759853 title "Fake Node Attacks on Graph Convolutional Networks" @default.
- W4312759853 doi "https://doi.org/10.47852/bonviewjcce2202321" @default.
- W4312759853 hasPublicationYear "2022" @default.
- W4312759853 type Work @default.
- W4312759853 citedByCount "3" @default.
- W4312759853 countsByYear W43127598532023 @default.
- W4312759853 crossrefType "journal-article" @default.
- W4312759853 hasAuthorship W4312759853A5000534132 @default.
- W4312759853 hasAuthorship W4312759853A5010841999 @default.
- W4312759853 hasAuthorship W4312759853A5046791174 @default.
- W4312759853 hasAuthorship W4312759853A5073670556 @default.
- W4312759853 hasAuthorship W4312759853A5088633029 @default.
- W4312759853 hasBestOaLocation W43127598531 @default.
- W4312759853 hasConcept C104317684 @default.
- W4312759853 hasConcept C11413529 @default.
- W4312759853 hasConcept C127413603 @default.
- W4312759853 hasConcept C132525143 @default.
- W4312759853 hasConcept C154945302 @default.
- W4312759853 hasConcept C185592680 @default.
- W4312759853 hasConcept C2779803651 @default.
- W4312759853 hasConcept C31258907 @default.
- W4312759853 hasConcept C37736160 @default.
- W4312759853 hasConcept C38652104 @default.
- W4312759853 hasConcept C41008148 @default.
- W4312759853 hasConcept C51823790 @default.
- W4312759853 hasConcept C55493867 @default.
- W4312759853 hasConcept C62611344 @default.
- W4312759853 hasConcept C63479239 @default.
- W4312759853 hasConcept C65856478 @default.
- W4312759853 hasConcept C66938386 @default.
- W4312759853 hasConcept C76155785 @default.
- W4312759853 hasConcept C80444323 @default.
- W4312759853 hasConcept C94915269 @default.
- W4312759853 hasConceptScore W4312759853C104317684 @default.
- W4312759853 hasConceptScore W4312759853C11413529 @default.
- W4312759853 hasConceptScore W4312759853C127413603 @default.
- W4312759853 hasConceptScore W4312759853C132525143 @default.
- W4312759853 hasConceptScore W4312759853C154945302 @default.
- W4312759853 hasConceptScore W4312759853C185592680 @default.
- W4312759853 hasConceptScore W4312759853C2779803651 @default.
- W4312759853 hasConceptScore W4312759853C31258907 @default.
- W4312759853 hasConceptScore W4312759853C37736160 @default.
- W4312759853 hasConceptScore W4312759853C38652104 @default.
- W4312759853 hasConceptScore W4312759853C41008148 @default.
- W4312759853 hasConceptScore W4312759853C51823790 @default.
- W4312759853 hasConceptScore W4312759853C55493867 @default.
- W4312759853 hasConceptScore W4312759853C62611344 @default.
- W4312759853 hasConceptScore W4312759853C63479239 @default.
- W4312759853 hasConceptScore W4312759853C65856478 @default.
- W4312759853 hasConceptScore W4312759853C66938386 @default.
- W4312759853 hasConceptScore W4312759853C76155785 @default.
- W4312759853 hasConceptScore W4312759853C80444323 @default.
- W4312759853 hasConceptScore W4312759853C94915269 @default.
- W4312759853 hasLocation W43127598531 @default.
- W4312759853 hasLocation W43127598532 @default.
- W4312759853 hasOpenAccess W4312759853 @default.
- W4312759853 hasPrimaryLocation W43127598531 @default.
- W4312759853 hasRelatedWork W2891262038 @default.
- W4312759853 hasRelatedWork W2945796017 @default.
- W4312759853 hasRelatedWork W2963870211 @default.
- W4312759853 hasRelatedWork W2964583308 @default.
- W4312759853 hasRelatedWork W2998122931 @default.
- W4312759853 hasRelatedWork W3138091863 @default.
- W4312759853 hasRelatedWork W3178864044 @default.
- W4312759853 hasRelatedWork W4308004328 @default.
- W4312759853 hasRelatedWork W4385765254 @default.
- W4312759853 hasRelatedWork W4386065769 @default.
- W4312759853 isParatext "false" @default.
- W4312759853 isRetracted "false" @default.
- W4312759853 workType "article" @default.