FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312728416> ?p ?o ?g. }

• W4312728416 endingPage "739" @default.
• W4312728416 startingPage "720" @default.
• W4312728416 abstract "Edge intelligence has arisen as a promising computing paradigm for supporting miscellaneous smart applications that rely on machine learning techniques. While the community has extensively investigated multi-tier edge deployment for traditional deep learning models (e.g. CNNs, RNNs), the emerging Graph Neural Networks (GNNs) are still under exploration, presenting a stark disparity to its broad edge adoptions such as traffic flow forecasting and location-based social recommendation. To bridge this gap, this paper formally studies the cost optimization for distributed GNN processing over a multi-tier heterogeneous edge network. We build a comprehensive modeling framework that can capture a variety of different cost factors, based on which we formulate a cost-efficient graph layout optimization problem that is proved to be NP-hard. Instead of trivially applying traditional data placement wisdom, we theoretically reveal the structural property of quadratic submodularity implicated in GNN’s unique computing pattern, which motivates our design of an efficient iterative solution exploiting graph cuts. Rigorous analysis shows that it provides parameterized constant approximation ratio, guaranteed convergence, and exact feasibility. To tackle potential graph topological evolution in GNN processing, we further devise an incremental update strategy and an adaptive scheduling algorithm for lightweight dynamic layout optimization. Evaluations with real-world datasets and various GNN benchmarks demonstrate that our approach achieves superior performance over de facto baselines with more than 95.8% cost reduction in a fast convergence speed." @default.
• W4312728416 created "2023-01-05" @default.
• W4312728416 creator A5015497946 @default.
• W4312728416 creator A5019752881 @default.
• W4312728416 creator A5053629236 @default.
• W4312728416 creator A5055161955 @default.
• W4312728416 creator A5071477872 @default.
• W4312728416 creator A5078909773 @default.
• W4312728416 date "2023-03-01" @default.
• W4312728416 modified "2023-10-14" @default.
• W4312728416 title "GNN at the Edge: Cost-Efficient Graph Neural Network Processing Over Distributed Edge Servers" @default.
• W4312728416 cites W1578015699 @default.
• W4312728416 cites W1983383464 @default.
• W4312728416 cites W2044033476 @default.
• W4312728416 cites W2045271686 @default.
• W4312728416 cites W2064058256 @default.
• W4312728416 cites W2090827575 @default.
• W4312728416 cites W2101309634 @default.
• W4312728416 cites W2139041001 @default.
• W4312728416 cites W2150593711 @default.
• W4312728416 cites W2157054705 @default.
• W4312728416 cites W2162528816 @default.
• W4312728416 cites W2416799949 @default.
• W4312728416 cites W2500139799 @default.
• W4312728416 cites W2612193523 @default.
• W4312728416 cites W2809740924 @default.
• W4312728416 cites W2883863832 @default.
• W4312728416 cites W2892341857 @default.
• W4312728416 cites W2896180420 @default.
• W4312728416 cites W2903798343 @default.
• W4312728416 cites W2903871660 @default.
• W4312728416 cites W2910151919 @default.
• W4312728416 cites W2919115771 @default.
• W4312728416 cites W2950865323 @default.
• W4312728416 cites W2955979265 @default.
• W4312728416 cites W2962883027 @default.
• W4312728416 cites W2964571482 @default.
• W4312728416 cites W2970929262 @default.
• W4312728416 cites W2979679572 @default.
• W4312728416 cites W2980856918 @default.
• W4312728416 cites W2982579831 @default.
• W4312728416 cites W3010622806 @default.
• W4312728416 cites W3011667710 @default.
• W4312728416 cites W3012562343 @default.
• W4312728416 cites W3014252079 @default.
• W4312728416 cites W3015689570 @default.
• W4312728416 cites W3017228913 @default.
• W4312728416 cites W3024560045 @default.
• W4312728416 cites W3034326350 @default.
• W4312728416 cites W3037702327 @default.
• W4312728416 cites W3040329870 @default.
• W4312728416 cites W3042370959 @default.
• W4312728416 cites W3049640275 @default.
• W4312728416 cites W3068123808 @default.
• W4312728416 cites W3083006614 @default.
• W4312728416 cites W3086238199 @default.
• W4312728416 cites W3092080090 @default.
• W4312728416 cites W3093741743 @default.
• W4312728416 cites W3099825604 @default.
• W4312728416 cites W3110777925 @default.
• W4312728416 cites W3123909522 @default.
• W4312728416 cites W3126230197 @default.
• W4312728416 cites W3130421533 @default.
• W4312728416 cites W3132522414 @default.
• W4312728416 cites W3133028082 @default.
• W4312728416 cites W3137762252 @default.
• W4312728416 cites W3154818219 @default.
• W4312728416 cites W3157805807 @default.
• W4312728416 cites W3158027451 @default.
• W4312728416 cites W3159109662 @default.
• W4312728416 cites W3159953606 @default.
• W4312728416 cites W3173386267 @default.
• W4312728416 cites W3180608480 @default.
• W4312728416 cites W3198787693 @default.
• W4312728416 cites W3211830369 @default.
• W4312728416 cites W4205645269 @default.
• W4312728416 cites W4206495448 @default.
• W4312728416 cites W4214578516 @default.
• W4312728416 cites W4224322855 @default.
• W4312728416 cites W4236099117 @default.
• W4312728416 cites W4281686206 @default.
• W4312728416 cites W4287890958 @default.
• W4312728416 cites W4290945652 @default.
• W4312728416 cites W4292873626 @default.
• W4312728416 cites W4294106961 @default.
• W4312728416 cites W4311080353 @default.
• W4312728416 doi "https://doi.org/10.1109/jsac.2022.3229422" @default.
• W4312728416 hasPublicationYear "2023" @default.
• W4312728416 type Work @default.
• W4312728416 citedByCount "0" @default.
• W4312728416 crossrefType "journal-article" @default.
• W4312728416 hasAuthorship W4312728416A5015497946 @default.
• W4312728416 hasAuthorship W4312728416A5019752881 @default.
• W4312728416 hasAuthorship W4312728416A5053629236 @default.
• W4312728416 hasAuthorship W4312728416A5055161955 @default.
• W4312728416 hasAuthorship W4312728416A5071477872 @default.
• W4312728416 hasAuthorship W4312728416A5078909773 @default.
• W4312728416 hasBestOaLocation W43127284162 @default.

• next