Matches in SemOpenAlex for { <https://semopenalex.org/work/W3193669819> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W3193669819 endingPage "e693" @default.
- W3193669819 startingPage "e693" @default.
- W3193669819 abstract "In recent years, graph convolutional networks (GCNs) have emerged rapidly due to their excellent performance in graph data processing. However, recent researches show that GCNs are vulnerable to adversarial attacks. An attacker can maliciously modify edges or nodes of the graph to mislead the model’s classification of the target nodes, or even cause a degradation of the model’s overall classification performance. In this paper, we first propose a black-box adversarial attack framework based on derivative-free optimization (DFO) to generate graph adversarial examples without using gradient and apply advanced DFO algorithms conveniently. Second, we implement a direct attack algorithm (DFDA) using the Nevergrad library based on the framework. Additionally, we overcome the problem of large search space by redesigning the perturbation vector using constraint size. Finally, we conducted a series of experiments on different datasets and parameters. The results show that DFDA outperforms Nettack in most cases, and it can achieve an average attack success rate of more than 95% on the Cora dataset when perturbing at most eight edges. This demonstrates that our framework can fully exploit the potential of DFO methods in node classification adversarial attacks." @default.
- W3193669819 created "2021-08-30" @default.
- W3193669819 creator A5001093757 @default.
- W3193669819 creator A5018688345 @default.
- W3193669819 date "2021-08-24" @default.
- W3193669819 modified "2023-10-16" @default.
- W3193669819 title "Derivative-free optimization adversarial attacks for graph convolutional networks" @default.
- W3193669819 cites W181733065 @default.
- W3193669819 cites W2116341502 @default.
- W3193669819 cites W2804452021 @default.
- W3193669819 cites W2899157217 @default.
- W3193669819 cites W2963613699 @default.
- W3193669819 doi "https://doi.org/10.7717/peerj-cs.693" @default.
- W3193669819 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/8409335" @default.
- W3193669819 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34541312" @default.
- W3193669819 hasPublicationYear "2021" @default.
- W3193669819 type Work @default.
- W3193669819 sameAs 3193669819 @default.
- W3193669819 citedByCount "1" @default.
- W3193669819 countsByYear W31936698192022 @default.
- W3193669819 crossrefType "journal-article" @default.
- W3193669819 hasAuthorship W3193669819A5001093757 @default.
- W3193669819 hasAuthorship W3193669819A5018688345 @default.
- W3193669819 hasBestOaLocation W31936698191 @default.
- W3193669819 hasConcept C132525143 @default.
- W3193669819 hasConcept C154945302 @default.
- W3193669819 hasConcept C165696696 @default.
- W3193669819 hasConcept C37736160 @default.
- W3193669819 hasConcept C38652104 @default.
- W3193669819 hasConcept C41008148 @default.
- W3193669819 hasConcept C80444323 @default.
- W3193669819 hasConceptScore W3193669819C132525143 @default.
- W3193669819 hasConceptScore W3193669819C154945302 @default.
- W3193669819 hasConceptScore W3193669819C165696696 @default.
- W3193669819 hasConceptScore W3193669819C37736160 @default.
- W3193669819 hasConceptScore W3193669819C38652104 @default.
- W3193669819 hasConceptScore W3193669819C41008148 @default.
- W3193669819 hasConceptScore W3193669819C80444323 @default.
- W3193669819 hasFunder F4320321001 @default.
- W3193669819 hasLocation W31936698191 @default.
- W3193669819 hasLocation W31936698192 @default.
- W3193669819 hasLocation W31936698193 @default.
- W3193669819 hasOpenAccess W3193669819 @default.
- W3193669819 hasPrimaryLocation W31936698191 @default.
- W3193669819 hasRelatedWork W1507793640 @default.
- W3193669819 hasRelatedWork W2945796017 @default.
- W3193669819 hasRelatedWork W2952419077 @default.
- W3193669819 hasRelatedWork W2953770707 @default.
- W3193669819 hasRelatedWork W2964583308 @default.
- W3193669819 hasRelatedWork W3034334914 @default.
- W3193669819 hasRelatedWork W4285252891 @default.
- W3193669819 hasRelatedWork W4288293779 @default.
- W3193669819 hasRelatedWork W4297899248 @default.
- W3193669819 hasRelatedWork W4310879833 @default.
- W3193669819 hasVolume "7" @default.
- W3193669819 isParatext "false" @default.
- W3193669819 isRetracted "false" @default.
- W3193669819 magId "3193669819" @default.
- W3193669819 workType "article" @default.