FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2887307029> ?p ?o ?g. }

• W2887307029 abstract "Convolutional neural networks (CNNs) have found extensive applications in practice. However, weight/activation's sparsity and different data precision requirements across layers lead to a large amount of redundant computations. In this paper, we propose an efficient architecture for CNNs, named Folded Precision-Adjustable Processor (FPAP), to skip those unnecessary computations with ease. Computations are folded in the following two aspects to achieve efficient computing. On one hand, the dominant multiply-and-add (MAC) operations are performed bit-serially based on a bit-pair encoding algorithm so that the FPAP can adapt to different numerical precisions without using multipliers with long data width. On the other hand, a 1-D convolution is undertaken by a multi-tap transposed finite impulse response (FIR) filter, which is folded into one tap so that computations involving zero activations and weights can be easily skipped. Equipped with the precision-adjustable MAC unit and the folded FIR filter structure, a well-designed array architecture, consisting of many identical processing elements is developed, which is scalable for different throughput requirements and highly flexible for different numerical precisions. Besides, a novel genetic algorithm based kernel reallocation scheme is introduced to mitigate the load imbalance issue. Our synthesis results demonstrate that the proposed FPAP can significantly reduce the logic complexity and the critical path over the corresponding unfolded design, which only delivers slightly higher throughput when processing sparse and compact models. Our experiments also show that FPAP can scale its energy efficiency from 1.01TOP/s/W to 6.26TOP/s/W under 90nm CMOS technology when different data precisions are used." @default.
• W2887307029 created "2018-08-22" @default.
• W2887307029 creator A5019940239 @default.
• W2887307029 creator A5047445405 @default.
• W2887307029 creator A5083351215 @default.
• W2887307029 date "2018-07-01" @default.
• W2887307029 modified "2023-10-08" @default.
• W2887307029 title "FPAP: A Folded Architecture for Efficient Computing of Convolutional Neural Networks" @default.
• W2887307029 cites W2047094503 @default.
• W2887307029 cites W2048266589 @default.
• W2887307029 cites W2163605009 @default.
• W2887307029 cites W2286365479 @default.
• W2887307029 cites W2300242332 @default.
• W2887307029 cites W2405920868 @default.
• W2887307029 cites W2516141709 @default.
• W2887307029 cites W2560017826 @default.
• W2887307029 cites W2563587242 @default.
• W2887307029 cites W2565851976 @default.
• W2887307029 cites W2613119772 @default.
• W2887307029 cites W2625457103 @default.
• W2887307029 cites W2962993358 @default.
• W2887307029 cites W2963114950 @default.
• W2887307029 cites W2964299589 @default.
• W2887307029 cites W3024621361 @default.
• W2887307029 doi "https://doi.org/10.1109/isvlsi.2018.00098" @default.
• W2887307029 hasPublicationYear "2018" @default.
• W2887307029 type Work @default.
• W2887307029 sameAs 2887307029 @default.
• W2887307029 citedByCount "2" @default.
• W2887307029 countsByYear W28873070292020 @default.
• W2887307029 countsByYear W28873070292021 @default.
• W2887307029 crossrefType "proceedings-article" @default.
• W2887307029 hasAuthorship W2887307029A5019940239 @default.
• W2887307029 hasAuthorship W2887307029A5047445405 @default.
• W2887307029 hasAuthorship W2887307029A5083351215 @default.
• W2887307029 hasConcept C106131492 @default.
• W2887307029 hasConcept C113775141 @default.
• W2887307029 hasConcept C11413529 @default.
• W2887307029 hasConcept C114614502 @default.
• W2887307029 hasConcept C125411270 @default.
• W2887307029 hasConcept C154945302 @default.
• W2887307029 hasConcept C157764524 @default.
• W2887307029 hasConcept C173608175 @default.
• W2887307029 hasConcept C198386975 @default.
• W2887307029 hasConcept C31972630 @default.
• W2887307029 hasConcept C33923547 @default.
• W2887307029 hasConcept C41008148 @default.
• W2887307029 hasConcept C45347329 @default.
• W2887307029 hasConcept C45374587 @default.
• W2887307029 hasConcept C459310 @default.
• W2887307029 hasConcept C48044578 @default.
• W2887307029 hasConcept C50644808 @default.
• W2887307029 hasConcept C555944384 @default.
• W2887307029 hasConcept C74193536 @default.
• W2887307029 hasConcept C76155785 @default.
• W2887307029 hasConcept C77088390 @default.
• W2887307029 hasConcept C81363708 @default.
• W2887307029 hasConcept C9390403 @default.
• W2887307029 hasConceptScore W2887307029C106131492 @default.
• W2887307029 hasConceptScore W2887307029C113775141 @default.
• W2887307029 hasConceptScore W2887307029C11413529 @default.
• W2887307029 hasConceptScore W2887307029C114614502 @default.
• W2887307029 hasConceptScore W2887307029C125411270 @default.
• W2887307029 hasConceptScore W2887307029C154945302 @default.
• W2887307029 hasConceptScore W2887307029C157764524 @default.
• W2887307029 hasConceptScore W2887307029C173608175 @default.
• W2887307029 hasConceptScore W2887307029C198386975 @default.
• W2887307029 hasConceptScore W2887307029C31972630 @default.
• W2887307029 hasConceptScore W2887307029C33923547 @default.
• W2887307029 hasConceptScore W2887307029C41008148 @default.
• W2887307029 hasConceptScore W2887307029C45347329 @default.
• W2887307029 hasConceptScore W2887307029C45374587 @default.
• W2887307029 hasConceptScore W2887307029C459310 @default.
• W2887307029 hasConceptScore W2887307029C48044578 @default.
• W2887307029 hasConceptScore W2887307029C50644808 @default.
• W2887307029 hasConceptScore W2887307029C555944384 @default.
• W2887307029 hasConceptScore W2887307029C74193536 @default.
• W2887307029 hasConceptScore W2887307029C76155785 @default.
• W2887307029 hasConceptScore W2887307029C77088390 @default.
• W2887307029 hasConceptScore W2887307029C81363708 @default.
• W2887307029 hasConceptScore W2887307029C9390403 @default.
• W2887307029 hasLocation W28873070291 @default.
• W2887307029 hasOpenAccess W2887307029 @default.
• W2887307029 hasPrimaryLocation W28873070291 @default.
• W2887307029 hasRelatedWork W1502254990 @default.
• W2887307029 hasRelatedWork W1574964655 @default.
• W2887307029 hasRelatedWork W2014621955 @default.
• W2887307029 hasRelatedWork W2557842770 @default.
• W2887307029 hasRelatedWork W2732805655 @default.
• W2887307029 hasRelatedWork W2734851905 @default.
• W2887307029 hasRelatedWork W2767737961 @default.
• W2887307029 hasRelatedWork W2831313928 @default.
• W2887307029 hasRelatedWork W2905184098 @default.
• W2887307029 hasRelatedWork W3005712428 @default.
• W2887307029 hasRelatedWork W3128905096 @default.
• W2887307029 hasRelatedWork W3134750800 @default.
• W2887307029 hasRelatedWork W3157663361 @default.
• W2887307029 hasRelatedWork W3159489986 @default.
• W2887307029 hasRelatedWork W3210059290 @default.
• W2887307029 hasRelatedWork W2468121480 @default.

• next