Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285284078> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4285284078 endingPage "15524" @default.
- W4285284078 startingPage "15511" @default.
- W4285284078 abstract "Convolutional neural networks (CNNS) enable machines to view the world as humans and become increasing prevalent for Internet of Things (IoT) applications. Instead of streaming the raw data to the cloud and executing CNN inference remotely, it would be very attractive to use local IoT devices to process as it enables IoT applications with independent decision-making ability. Since a single IoT device can hardly match the requirements of the CNN inference, especially for time-sensitive and high-accuracy tasks, the distributed <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>in-situ</i> CNN inference becomes a potential solution. However, because of the inherently tightly coupled structure of existing CNN models, it is difficult to distribute the inference efficiently. In this article, we enhance the distributed <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>in-situ</i> CNN inference in the IoT. We fundamentally reduce the communication overhead of distributed CNN inference by designing new loosely coupled structure (LCS). Experimental results demonstrate that LCS achieves the leading performance compared with other popular structures. Next, based on the LCS, we customize the partitioning method to reduce the synchronization points and design the decentralized asynchronous method to optimize communication in each synchronization point. To evaluate the effectiveness, we build a prototype system. When the number of IoT devices increases from 1 to 4, our system accelerates by up to <inline-formula xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink> <tex-math notation=LaTeX>$3.85times $ </tex-math></inline-formula> and reduces the memory footprint in each device by 70% with achieving a competitive accuracy and significantly outperforming other approaches." @default.
- W4285284078 created "2022-07-14" @default.
- W4285284078 creator A5000151117 @default.
- W4285284078 creator A5024085713 @default.
- W4285284078 creator A5034484688 @default.
- W4285284078 creator A5056096237 @default.
- W4285284078 creator A5059440572 @default.
- W4285284078 date "2022-09-01" @default.
- W4285284078 modified "2023-09-28" @default.
- W4285284078 title "Enhancing Distributed In-Situ CNN Inference in the Internet of Things" @default.
- W4285284078 cites W2117539524 @default.
- W4285284078 cites W2194775991 @default.
- W4285284078 cites W2468676150 @default.
- W4285284078 cites W2568772110 @default.
- W4285284078 cites W2680270903 @default.
- W4285284078 cites W2806663193 @default.
- W4285284078 cites W2891123429 @default.
- W4285284078 cites W2897303255 @default.
- W4285284078 cites W2905207581 @default.
- W4285284078 cites W2922395136 @default.
- W4285284078 cites W2945129842 @default.
- W4285284078 cites W2946719914 @default.
- W4285284078 cites W2963037989 @default.
- W4285284078 cites W2963125010 @default.
- W4285284078 cites W2963163009 @default.
- W4285284078 cites W2963363373 @default.
- W4285284078 cites W2963993763 @default.
- W4285284078 cites W2990958518 @default.
- W4285284078 cites W3004061291 @default.
- W4285284078 cites W3035038672 @default.
- W4285284078 cites W3104849992 @default.
- W4285284078 cites W3105966348 @default.
- W4285284078 cites W3108835732 @default.
- W4285284078 cites W3117065193 @default.
- W4285284078 cites W3215915968 @default.
- W4285284078 doi "https://doi.org/10.1109/jiot.2022.3176408" @default.
- W4285284078 hasPublicationYear "2022" @default.
- W4285284078 type Work @default.
- W4285284078 citedByCount "0" @default.
- W4285284078 crossrefType "journal-article" @default.
- W4285284078 hasAuthorship W4285284078A5000151117 @default.
- W4285284078 hasAuthorship W4285284078A5024085713 @default.
- W4285284078 hasAuthorship W4285284078A5034484688 @default.
- W4285284078 hasAuthorship W4285284078A5056096237 @default.
- W4285284078 hasAuthorship W4285284078A5059440572 @default.
- W4285284078 hasConcept C111919701 @default.
- W4285284078 hasConcept C119857082 @default.
- W4285284078 hasConcept C120314980 @default.
- W4285284078 hasConcept C127162648 @default.
- W4285284078 hasConcept C149635348 @default.
- W4285284078 hasConcept C151319957 @default.
- W4285284078 hasConcept C154945302 @default.
- W4285284078 hasConcept C2776214188 @default.
- W4285284078 hasConcept C2778562939 @default.
- W4285284078 hasConcept C2779960059 @default.
- W4285284078 hasConcept C31258907 @default.
- W4285284078 hasConcept C41008148 @default.
- W4285284078 hasConcept C81363708 @default.
- W4285284078 hasConcept C81860439 @default.
- W4285284078 hasConcept C98045186 @default.
- W4285284078 hasConceptScore W4285284078C111919701 @default.
- W4285284078 hasConceptScore W4285284078C119857082 @default.
- W4285284078 hasConceptScore W4285284078C120314980 @default.
- W4285284078 hasConceptScore W4285284078C127162648 @default.
- W4285284078 hasConceptScore W4285284078C149635348 @default.
- W4285284078 hasConceptScore W4285284078C151319957 @default.
- W4285284078 hasConceptScore W4285284078C154945302 @default.
- W4285284078 hasConceptScore W4285284078C2776214188 @default.
- W4285284078 hasConceptScore W4285284078C2778562939 @default.
- W4285284078 hasConceptScore W4285284078C2779960059 @default.
- W4285284078 hasConceptScore W4285284078C31258907 @default.
- W4285284078 hasConceptScore W4285284078C41008148 @default.
- W4285284078 hasConceptScore W4285284078C81363708 @default.
- W4285284078 hasConceptScore W4285284078C81860439 @default.
- W4285284078 hasConceptScore W4285284078C98045186 @default.
- W4285284078 hasFunder F4320321001 @default.
- W4285284078 hasFunder F4320335777 @default.
- W4285284078 hasIssue "17" @default.
- W4285284078 hasLocation W42852840781 @default.
- W4285284078 hasOpenAccess W4285284078 @default.
- W4285284078 hasPrimaryLocation W42852840781 @default.
- W4285284078 hasRelatedWork W1561560534 @default.
- W4285284078 hasRelatedWork W2092071486 @default.
- W4285284078 hasRelatedWork W2131630752 @default.
- W4285284078 hasRelatedWork W2147711412 @default.
- W4285284078 hasRelatedWork W2313989154 @default.
- W4285284078 hasRelatedWork W2350861899 @default.
- W4285284078 hasRelatedWork W2364921833 @default.
- W4285284078 hasRelatedWork W2391167130 @default.
- W4285284078 hasRelatedWork W3027997911 @default.
- W4285284078 hasRelatedWork W4287776258 @default.
- W4285284078 hasVolume "9" @default.
- W4285284078 isParatext "false" @default.
- W4285284078 isRetracted "false" @default.
- W4285284078 workType "article" @default.