Matches in SemOpenAlex for { <https://semopenalex.org/work/W2949854114> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W2949854114 abstract "Graph neural networks (GNNs) which apply the deep neural networks to graph data have achieved significant performance for the task of semi-supervised node classification. However, only few work has addressed the adversarial robustness of GNNs. In this paper, we first present a novel gradient-based attack method that facilitates the difficulty of tackling discrete graph data. When comparing to current adversarial attacks on GNNs, the results show that by only perturbing a small number of edge perturbations, including addition and deletion, our optimization-based attack can lead to a noticeable decrease in classification performance. Moreover, leveraging our gradient-based attack, we propose the first optimization-based adversarial training for GNNs. Our method yields higher robustness against both different gradient based and greedy attack methods without sacrificing classification accuracy on original graph." @default.
- W2949854114 created "2019-06-27" @default.
- W2949854114 creator A5002976916 @default.
- W2949854114 creator A5007306953 @default.
- W2949854114 creator A5011173717 @default.
- W2949854114 creator A5016900183 @default.
- W2949854114 creator A5043582832 @default.
- W2949854114 creator A5050344371 @default.
- W2949854114 creator A5062177930 @default.
- W2949854114 date "2019-06-10" @default.
- W2949854114 modified "2023-09-25" @default.
- W2949854114 title "Topology Attack and Defense for Graph Neural Networks: An Optimization Perspective" @default.
- W2949854114 cites W2069812095 @default.
- W2949854114 cites W2257439149 @default.
- W2949854114 cites W2296319761 @default.
- W2949854114 cites W2519887557 @default.
- W2949854114 cites W2640329709 @default.
- W2949854114 cites W2766453196 @default.
- W2949854114 cites W2803429660 @default.
- W2949854114 cites W2803678876 @default.
- W2949854114 cites W2803831897 @default.
- W2949854114 cites W2887906331 @default.
- W2949854114 cites W2914953695 @default.
- W2949854114 cites W2934684307 @default.
- W2949854114 cites W2937268132 @default.
- W2949854114 cites W2946297661 @default.
- W2949854114 cites W2950448984 @default.
- W2949854114 cites W2962711740 @default.
- W2949854114 cites W2962730916 @default.
- W2949854114 cites W2962995403 @default.
- W2949854114 cites W2963207607 @default.
- W2949854114 cites W2963389226 @default.
- W2949854114 cites W2963612069 @default.
- W2949854114 cites W2963857521 @default.
- W2949854114 cites W2964301649 @default.
- W2949854114 cites W3104097132 @default.
- W2949854114 doi "https://doi.org/10.48550/arxiv.1906.04214" @default.
- W2949854114 hasPublicationYear "2019" @default.
- W2949854114 type Work @default.
- W2949854114 sameAs 2949854114 @default.
- W2949854114 citedByCount "49" @default.
- W2949854114 countsByYear W29498541142019 @default.
- W2949854114 countsByYear W29498541142020 @default.
- W2949854114 countsByYear W29498541142021 @default.
- W2949854114 crossrefType "posted-content" @default.
- W2949854114 hasAuthorship W2949854114A5002976916 @default.
- W2949854114 hasAuthorship W2949854114A5007306953 @default.
- W2949854114 hasAuthorship W2949854114A5011173717 @default.
- W2949854114 hasAuthorship W2949854114A5016900183 @default.
- W2949854114 hasAuthorship W2949854114A5043582832 @default.
- W2949854114 hasAuthorship W2949854114A5050344371 @default.
- W2949854114 hasAuthorship W2949854114A5062177930 @default.
- W2949854114 hasBestOaLocation W29498541141 @default.
- W2949854114 hasConcept C104317684 @default.
- W2949854114 hasConcept C119857082 @default.
- W2949854114 hasConcept C126255220 @default.
- W2949854114 hasConcept C132525143 @default.
- W2949854114 hasConcept C154945302 @default.
- W2949854114 hasConcept C185592680 @default.
- W2949854114 hasConcept C193254401 @default.
- W2949854114 hasConcept C2984842247 @default.
- W2949854114 hasConcept C33923547 @default.
- W2949854114 hasConcept C37736160 @default.
- W2949854114 hasConcept C41008148 @default.
- W2949854114 hasConcept C50644808 @default.
- W2949854114 hasConcept C55493867 @default.
- W2949854114 hasConcept C63479239 @default.
- W2949854114 hasConcept C80444323 @default.
- W2949854114 hasConceptScore W2949854114C104317684 @default.
- W2949854114 hasConceptScore W2949854114C119857082 @default.
- W2949854114 hasConceptScore W2949854114C126255220 @default.
- W2949854114 hasConceptScore W2949854114C132525143 @default.
- W2949854114 hasConceptScore W2949854114C154945302 @default.
- W2949854114 hasConceptScore W2949854114C185592680 @default.
- W2949854114 hasConceptScore W2949854114C193254401 @default.
- W2949854114 hasConceptScore W2949854114C2984842247 @default.
- W2949854114 hasConceptScore W2949854114C33923547 @default.
- W2949854114 hasConceptScore W2949854114C37736160 @default.
- W2949854114 hasConceptScore W2949854114C41008148 @default.
- W2949854114 hasConceptScore W2949854114C50644808 @default.
- W2949854114 hasConceptScore W2949854114C55493867 @default.
- W2949854114 hasConceptScore W2949854114C63479239 @default.
- W2949854114 hasConceptScore W2949854114C80444323 @default.
- W2949854114 hasLocation W29498541141 @default.
- W2949854114 hasOpenAccess W2949854114 @default.
- W2949854114 hasPrimaryLocation W29498541141 @default.
- W2949854114 hasRelatedWork W2516574342 @default.
- W2949854114 hasRelatedWork W2610321374 @default.
- W2949854114 hasRelatedWork W2626387500 @default.
- W2949854114 hasRelatedWork W2952919291 @default.
- W2949854114 hasRelatedWork W2963857521 @default.
- W2949854114 hasRelatedWork W2967955854 @default.
- W2949854114 hasRelatedWork W3164364099 @default.
- W2949854114 hasRelatedWork W3164563754 @default.
- W2949854114 hasRelatedWork W3193857078 @default.
- W2949854114 hasRelatedWork W3111275546 @default.
- W2949854114 isParatext "false" @default.
- W2949854114 isRetracted "false" @default.
- W2949854114 magId "2949854114" @default.
- W2949854114 workType "article" @default.