FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285505364> ?p ?o ?g. }

• W4285505364 endingPage "24515" @default.
• W4285505364 startingPage "24506" @default.
• W4285505364 abstract "Convolutional neural networks (CNNs) have been applied in numerous Internet of Things (IoT) devices for multifarious downstream tasks. However, with the increasing amount of data on edge devices, CNNs can hardly complete some tasks in time with limited computing and storage resources. Recently, filter pruning has been regarded as an effective technique to compress and accelerate CNNs, but existing methods rarely prune CNNs from the perspective of compressing high-dimensional tensors. In this article, we propose a novel theory to find redundant information in 3-D tensors, namely, quantified similarity between feature maps (QSFM), and utilize this theory to guide the filter pruning procedure. We perform QSFM on data sets (CIFAR-10, CIFAR-100, and ILSVRC-12) and edge devices and demonstrate that the proposed method can find the redundant information in the neural networks effectively with comparable compression and tolerable drop of accuracy. Without any fine-tuning operation, QSFM can compress ResNet-56 on CIFAR-10 significantly (48.7% FLOPs and 57.9% parameters are reduced) with only a loss of 0.54% in the top-1 accuracy. For the practical application of edge devices, QSFM can accelerate MobileNet-V2 inference speed by 1.53 times with only a loss of 1.23% in the ILSVRC-12 top-1 accuracy." @default.
• W4285505364 created "2022-07-15" @default.
• W4285505364 creator A5004119465 @default.
• W4285505364 creator A5015416321 @default.
• W4285505364 creator A5015551771 @default.
• W4285505364 creator A5038443450 @default.
• W4285505364 creator A5071490313 @default.
• W4285505364 creator A5087445936 @default.
• W4285505364 creator A5089553750 @default.
• W4285505364 date "2022-12-01" @default.
• W4285505364 modified "2023-10-15" @default.
• W4285505364 title "QSFM: Model Pruning Based on Quantified Similarity Between Feature Maps for AI on Edge" @default.
• W4285505364 cites W2097117768 @default.
• W4285505364 cites W2117539524 @default.
• W4285505364 cites W2133665775 @default.
• W4285505364 cites W2194775991 @default.
• W4285505364 cites W2382313035 @default.
• W4285505364 cites W2416799949 @default.
• W4285505364 cites W2785457004 @default.
• W4285505364 cites W2886851211 @default.
• W4285505364 cites W2894581376 @default.
• W4285505364 cites W2924515500 @default.
• W4285505364 cites W2962851801 @default.
• W4285505364 cites W2963145730 @default.
• W4285505364 cites W2963163009 @default.
• W4285505364 cites W2963363373 @default.
• W4285505364 cites W2963382930 @default.
• W4285505364 cites W2963446712 @default.
• W4285505364 cites W2963576971 @default.
• W4285505364 cites W2963626527 @default.
• W4285505364 cites W2964266063 @default.
• W4285505364 cites W2971544482 @default.
• W4285505364 cites W3034251466 @default.
• W4285505364 cites W3034513523 @default.
• W4285505364 cites W3034971973 @default.
• W4285505364 cites W3035414587 @default.
• W4285505364 cites W3097692969 @default.
• W4285505364 cites W3104300620 @default.
• W4285505364 cites W3126796721 @default.
• W4285505364 cites W3172740058 @default.
• W4285505364 doi "https://doi.org/10.1109/jiot.2022.3190873" @default.
• W4285505364 hasPublicationYear "2022" @default.
• W4285505364 type Work @default.
• W4285505364 citedByCount "4" @default.
• W4285505364 countsByYear W42855053642022 @default.
• W4285505364 countsByYear W42855053642023 @default.
• W4285505364 crossrefType "journal-article" @default.
• W4285505364 hasAuthorship W4285505364A5004119465 @default.
• W4285505364 hasAuthorship W4285505364A5015416321 @default.
• W4285505364 hasAuthorship W4285505364A5015551771 @default.
• W4285505364 hasAuthorship W4285505364A5038443450 @default.
• W4285505364 hasAuthorship W4285505364A5071490313 @default.
• W4285505364 hasAuthorship W4285505364A5087445936 @default.
• W4285505364 hasAuthorship W4285505364A5089553750 @default.
• W4285505364 hasBestOaLocation W42855053642 @default.
• W4285505364 hasConcept C103278499 @default.
• W4285505364 hasConcept C106131492 @default.
• W4285505364 hasConcept C108010975 @default.
• W4285505364 hasConcept C111919701 @default.
• W4285505364 hasConcept C115961682 @default.
• W4285505364 hasConcept C124101348 @default.
• W4285505364 hasConcept C138236772 @default.
• W4285505364 hasConcept C138885662 @default.
• W4285505364 hasConcept C153180895 @default.
• W4285505364 hasConcept C154945302 @default.
• W4285505364 hasConcept C162307627 @default.
• W4285505364 hasConcept C173608175 @default.
• W4285505364 hasConcept C2776214188 @default.
• W4285505364 hasConcept C2776401178 @default.
• W4285505364 hasConcept C31972630 @default.
• W4285505364 hasConcept C3826847 @default.
• W4285505364 hasConcept C41008148 @default.
• W4285505364 hasConcept C41895202 @default.
• W4285505364 hasConcept C50644808 @default.
• W4285505364 hasConcept C6557445 @default.
• W4285505364 hasConcept C79974875 @default.
• W4285505364 hasConcept C81363708 @default.
• W4285505364 hasConcept C86803240 @default.
• W4285505364 hasConceptScore W4285505364C103278499 @default.
• W4285505364 hasConceptScore W4285505364C106131492 @default.
• W4285505364 hasConceptScore W4285505364C108010975 @default.
• W4285505364 hasConceptScore W4285505364C111919701 @default.
• W4285505364 hasConceptScore W4285505364C115961682 @default.
• W4285505364 hasConceptScore W4285505364C124101348 @default.
• W4285505364 hasConceptScore W4285505364C138236772 @default.
• W4285505364 hasConceptScore W4285505364C138885662 @default.
• W4285505364 hasConceptScore W4285505364C153180895 @default.
• W4285505364 hasConceptScore W4285505364C154945302 @default.
• W4285505364 hasConceptScore W4285505364C162307627 @default.
• W4285505364 hasConceptScore W4285505364C173608175 @default.
• W4285505364 hasConceptScore W4285505364C2776214188 @default.
• W4285505364 hasConceptScore W4285505364C2776401178 @default.
• W4285505364 hasConceptScore W4285505364C31972630 @default.
• W4285505364 hasConceptScore W4285505364C3826847 @default.
• W4285505364 hasConceptScore W4285505364C41008148 @default.
• W4285505364 hasConceptScore W4285505364C41895202 @default.
• W4285505364 hasConceptScore W4285505364C50644808 @default.
• W4285505364 hasConceptScore W4285505364C6557445 @default.

• next