FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204085787> ?p ?o ?g. }

• W3204085787 abstract "Graph is an important data representation ubiquitously existing in the real world. However, analyzing the graph data is computationally difficult due to its non-Euclidean nature. Graph embedding is a powerful tool to solve the graph analytics problem by transforming the graph data into low-dimensional vectors. These vectors could also be shared with third parties to gain additional insights of what is behind the data. While sharing graph embedding is intriguing, the associated privacy risks are unexplored. In this paper, we systematically investigate the information leakage of the graph embedding by mounting three inference attacks. First, we can successfully infer basic graph properties, such as the number of nodes, the number of edges, and graph density, of the target graph with up to 0.89 accuracy. Second, given a subgraph of interest and the graph embedding, we can determine with high confidence that whether the subgraph is contained in the target graph. For instance, we achieve 0.98 attack AUC on the DD dataset. Third, we propose a novel graph reconstruction attack that can reconstruct a graph that has similar graph structural statistics to the target graph. We further propose an effective defense mechanism based on graph embedding perturbation to mitigate the inference attacks without noticeable performance degradation for graph classification tasks. Our code is available at https://github.com/Zhangzhk0819/GNN-Embedding-Leaks." @default.
• W3204085787 created "2021-10-11" @default.
• W3204085787 creator A5004014878 @default.
• W3204085787 creator A5012718640 @default.
• W3204085787 creator A5044420496 @default.
• W3204085787 creator A5050781579 @default.
• W3204085787 creator A5071833362 @default.
• W3204085787 date "2021-10-06" @default.
• W3204085787 modified "2023-09-26" @default.
• W3204085787 title "Inference Attacks Against Graph Neural Networks" @default.
• W3204085787 cites W1473189865 @default.
• W3204085787 cites W1991539813 @default.
• W3204085787 cites W2011039300 @default.
• W3204085787 cites W2142498761 @default.
• W3204085787 cites W2535690855 @default.
• W3204085787 cites W2605258629 @default.
• W3204085787 cites W2612872092 @default.
• W3204085787 cites W2624431344 @default.
• W3204085787 cites W2752520290 @default.
• W3204085787 cites W2757528734 @default.
• W3204085787 cites W2789304371 @default.
• W3204085787 cites W2803831897 @default.
• W3204085787 cites W2809583854 @default.
• W3204085787 cites W2811124557 @default.
• W3204085787 cites W2908442265 @default.
• W3204085787 cites W2912240430 @default.
• W3204085787 cites W2945796017 @default.
• W3204085787 cites W2951055356 @default.
• W3204085787 cites W2954831790 @default.
• W3204085787 cites W2962711740 @default.
• W3204085787 cites W2962756421 @default.
• W3204085787 cites W2962767366 @default.
• W3204085787 cites W2962810718 @default.
• W3204085787 cites W2963456518 @default.
• W3204085787 cites W2963858333 @default.
• W3204085787 cites W2963920355 @default.
• W3204085787 cites W2963953172 @default.
• W3204085787 cites W2963984147 @default.
• W3204085787 cites W2964015378 @default.
• W3204085787 cites W2964108670 @default.
• W3204085787 cites W2964151798 @default.
• W3204085787 cites W2964283260 @default.
• W3204085787 cites W2964321699 @default.
• W3204085787 cites W2964583308 @default.
• W3204085787 cites W2964971928 @default.
• W3204085787 cites W2970482593 @default.
• W3204085787 cites W2970552844 @default.
• W3204085787 cites W2983140679 @default.
• W3204085787 cites W2994506732 @default.
• W3204085787 cites W2995525544 @default.
• W3204085787 cites W2995983896 @default.
• W3204085787 cites W2998122931 @default.
• W3204085787 cites W3007332492 @default.
• W3204085787 cites W3007623546 @default.
• W3204085787 cites W3007913141 @default.
• W3204085787 cites W3011667710 @default.
• W3204085787 cites W3013520104 @default.
• W3204085787 cites W3018424040 @default.
• W3204085787 cites W3035010690 @default.
• W3204085787 cites W3036148123 @default.
• W3204085787 cites W3042313988 @default.
• W3204085787 cites W3048684575 @default.
• W3204085787 cites W3087257704 @default.
• W3204085787 cites W3092574271 @default.
• W3204085787 cites W3094559034 @default.
• W3204085787 cites W3096738375 @default.
• W3204085787 cites W3104097132 @default.
• W3204085787 cites W3105626348 @default.
• W3204085787 cites W3116489338 @default.
• W3204085787 cites W3189551349 @default.
• W3204085787 cites W3214586949 @default.
• W3204085787 cites W3156148409 @default.
• W3204085787 doi "https://doi.org/10.48550/arxiv.2110.02631" @default.
• W3204085787 hasPublicationYear "2021" @default.
• W3204085787 type Work @default.
• W3204085787 sameAs 3204085787 @default.
• W3204085787 citedByCount "0" @default.
• W3204085787 crossrefType "posted-content" @default.
• W3204085787 hasAuthorship W3204085787A5004014878 @default.
• W3204085787 hasAuthorship W3204085787A5012718640 @default.
• W3204085787 hasAuthorship W3204085787A5044420496 @default.
• W3204085787 hasAuthorship W3204085787A5050781579 @default.
• W3204085787 hasAuthorship W3204085787A5071833362 @default.
• W3204085787 hasBestOaLocation W32040857871 @default.
• W3204085787 hasConcept C132525143 @default.
• W3204085787 hasConcept C147792647 @default.
• W3204085787 hasConcept C154945302 @default.
• W3204085787 hasConcept C168291704 @default.
• W3204085787 hasConcept C17169500 @default.
• W3204085787 hasConcept C176225458 @default.
• W3204085787 hasConcept C18819970 @default.
• W3204085787 hasConcept C203776342 @default.
• W3204085787 hasConcept C22149727 @default.
• W3204085787 hasConcept C2776214188 @default.
• W3204085787 hasConcept C36038622 @default.
• W3204085787 hasConcept C41008148 @default.
• W3204085787 hasConcept C64339825 @default.
• W3204085787 hasConcept C75564084 @default.
• W3204085787 hasConcept C80444323 @default.
• W3204085787 hasConceptScore W3204085787C132525143 @default.

• next