FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285257856> ?p ?o ?g. }

• W4285257856 endingPage "1" @default.
• W4285257856 startingPage "1" @default.
• W4285257856 abstract "Edge intelligence, as a prospective paradigm for accelerating DNN inference, is mostly implemented by model partitioning which inevitably incurs the large transmission overhead of DNN's intermediate data. A popular solution introduces multi-exit DNNs to reduce latency by enabling early exits. However, existing work ignores the correlation between exit settings and synergistic inference, causing incoordination of device-to-edge. To address this issue, this paper first investigates the bottlenecks of executing multi-exit DNNs in edge computing and builds a novel model for inference acceleration with exit selection, model partition, and resource allocation. To tackle the intractable coupling subproblems, we propose a Multi-exit DNN inference Acceleration framework based on Multi-dimensional Optimization (MAMO). In MAMO, the exit selection subproblem is first extracted from the original problem. Then, bidirectional dynamic programming is employed to determine the optimal exit setting for an arbitrary multi-exit DNN. Finally, based on the optimal exit setting, a DRL-based policy is developed to learn joint decisions of model partition and resource allocation. We deploy MAMO on a real-world testbed and evaluate its performance in various scenarios. Extensive experiments show that it can adapt to heterogeneous tasks and dynamic networks, and accelerate DNN inference by up to 13.7x compared with the state-of-the-art." @default.
• W4285257856 created "2022-07-14" @default.
• W4285257856 creator A5008008825 @default.
• W4285257856 creator A5030336292 @default.
• W4285257856 creator A5038125746 @default.
• W4285257856 creator A5056420341 @default.
• W4285257856 creator A5067124205 @default.
• W4285257856 creator A5067735672 @default.
• W4285257856 creator A5073492643 @default.
• W4285257856 creator A5086482012 @default.
• W4285257856 date "2022-01-01" @default.
• W4285257856 modified "2023-09-26" @default.
• W4285257856 title "Multi-exit DNN Inference Acceleration based on Multi-Dimensional Optimization for Edge Intelligence" @default.
• W4285257856 cites W2183341477 @default.
• W4285257856 cites W2194775991 @default.
• W4285257856 cites W2416799949 @default.
• W4285257856 cites W2554302513 @default.
• W4285257856 cites W2609285553 @default.
• W4285257856 cites W2896180420 @default.
• W4285257856 cites W2912213068 @default.
• W4285257856 cites W2920031528 @default.
• W4285257856 cites W2931743911 @default.
• W4285257856 cites W2949628230 @default.
• W4285257856 cites W2950865323 @default.
• W4285257856 cites W2960833983 @default.
• W4285257856 cites W2962677625 @default.
• W4285257856 cites W2962804345 @default.
• W4285257856 cites W2962883027 @default.
• W4285257856 cites W2964223234 @default.
• W4285257856 cites W2966313861 @default.
• W4285257856 cites W2980856918 @default.
• W4285257856 cites W2981812042 @default.
• W4285257856 cites W2981884310 @default.
• W4285257856 cites W2989368243 @default.
• W4285257856 cites W3034513523 @default.
• W4285257856 cites W3044591330 @default.
• W4285257856 cites W3047565185 @default.
• W4285257856 cites W3049640275 @default.
• W4285257856 cites W3081179222 @default.
• W4285257856 cites W3147954149 @default.
• W4285257856 cites W3191795420 @default.
• W4285257856 cites W3203452446 @default.
• W4285257856 cites W3208262036 @default.
• W4285257856 cites W4224314467 @default.
• W4285257856 cites W4236099117 @default.
• W4285257856 cites W4249516033 @default.
• W4285257856 doi "https://doi.org/10.1109/tmc.2022.3172402" @default.
• W4285257856 hasPublicationYear "2022" @default.
• W4285257856 type Work @default.
• W4285257856 citedByCount "8" @default.
• W4285257856 countsByYear W42852578562022 @default.
• W4285257856 countsByYear W42852578562023 @default.
• W4285257856 crossrefType "journal-article" @default.
• W4285257856 hasAuthorship W4285257856A5008008825 @default.
• W4285257856 hasAuthorship W4285257856A5030336292 @default.
• W4285257856 hasAuthorship W4285257856A5038125746 @default.
• W4285257856 hasAuthorship W4285257856A5056420341 @default.
• W4285257856 hasAuthorship W4285257856A5067124205 @default.
• W4285257856 hasAuthorship W4285257856A5067735672 @default.
• W4285257856 hasAuthorship W4285257856A5073492643 @default.
• W4285257856 hasAuthorship W4285257856A5086482012 @default.
• W4285257856 hasConcept C111919701 @default.
• W4285257856 hasConcept C11413529 @default.
• W4285257856 hasConcept C114614502 @default.
• W4285257856 hasConcept C117896860 @default.
• W4285257856 hasConcept C120314980 @default.
• W4285257856 hasConcept C121332964 @default.
• W4285257856 hasConcept C137836250 @default.
• W4285257856 hasConcept C138236772 @default.
• W4285257856 hasConcept C154945302 @default.
• W4285257856 hasConcept C162307627 @default.
• W4285257856 hasConcept C2776214188 @default.
• W4285257856 hasConcept C31258907 @default.
• W4285257856 hasConcept C31395832 @default.
• W4285257856 hasConcept C33923547 @default.
• W4285257856 hasConcept C41008148 @default.
• W4285257856 hasConcept C42812 @default.
• W4285257856 hasConcept C74650414 @default.
• W4285257856 hasConcept C76155785 @default.
• W4285257856 hasConcept C79974875 @default.
• W4285257856 hasConcept C82876162 @default.
• W4285257856 hasConceptScore W4285257856C111919701 @default.
• W4285257856 hasConceptScore W4285257856C11413529 @default.
• W4285257856 hasConceptScore W4285257856C114614502 @default.
• W4285257856 hasConceptScore W4285257856C117896860 @default.
• W4285257856 hasConceptScore W4285257856C120314980 @default.
• W4285257856 hasConceptScore W4285257856C121332964 @default.
• W4285257856 hasConceptScore W4285257856C137836250 @default.
• W4285257856 hasConceptScore W4285257856C138236772 @default.
• W4285257856 hasConceptScore W4285257856C154945302 @default.
• W4285257856 hasConceptScore W4285257856C162307627 @default.
• W4285257856 hasConceptScore W4285257856C2776214188 @default.
• W4285257856 hasConceptScore W4285257856C31258907 @default.
• W4285257856 hasConceptScore W4285257856C31395832 @default.
• W4285257856 hasConceptScore W4285257856C33923547 @default.
• W4285257856 hasConceptScore W4285257856C41008148 @default.
• W4285257856 hasConceptScore W4285257856C42812 @default.
• W4285257856 hasConceptScore W4285257856C74650414 @default.

• next