FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2565851976> ?p ?o ?g. }

• W2565851976 endingPage "12" @default.
• W2565851976 startingPage "1" @default.
• W2565851976 abstract "Neural networks (NNs) have been demonstrated to be useful in a broad range of applications such as image recognition, automatic translation and advertisement recommendation. State-of-the-art NNs are known to be both computationally and memory intensive, due to the ever-increasing deep structure, i.e., multiple layers with massive neurons and connections (i.e., synapses). Sparse neural networks have emerged as an effective solution to reduce the amount of computation and memory required. Though existing NN accelerators are able to efficiently process dense and regular networks, they cannot benefit from the reduction of synaptic weights. In this paper, we propose a novel accelerator, Cambricon-X, to exploit the sparsity and irregularity of NN models for increased efficiency. The proposed accelerator features a PE-based architecture consisting of multiple Processing Elements (PE). An Indexing Module (IM) efficiently selects and transfers needed neurons to connected PEs with reduced bandwidth requirement, while each PE stores irregular and compressed synapses for local computation in an asynchronous fashion. With 16 PEs, our accelerator is able to achieve at most 544 GOP/s in a small form factor (6.38 mm2 and 954 mW at 65 nm). Experimental results over a number of representative sparse networks show that our accelerator achieves, on average, 7.23x speedup and 6.43x energy saving against the state-of-the-art NN accelerator." @default.
• W2565851976 created "2017-01-06" @default.
• W2565851976 creator A5014715855 @default.
• W2565851976 creator A5033237869 @default.
• W2565851976 creator A5034758732 @default.
• W2565851976 creator A5036492388 @default.
• W2565851976 creator A5056949827 @default.
• W2565851976 creator A5063724348 @default.
• W2565851976 creator A5072861069 @default.
• W2565851976 creator A5081598354 @default.
• W2565851976 date "2016-10-15" @default.
• W2565851976 modified "2023-09-23" @default.
• W2565851976 title "Cambricon-x: an accelerator for sparse neural networks" @default.
• W2565851976 cites W1891025735 @default.
• W2565851976 cites W1964471912 @default.
• W2565851976 cites W1990315422 @default.
• W2565851976 cites W2009832130 @default.
• W2565851976 cites W2010069327 @default.
• W2565851976 cites W2013237284 @default.
• W2565851976 cites W2014807599 @default.
• W2565851976 cites W2042876290 @default.
• W2565851976 cites W2043754576 @default.
• W2565851976 cites W2044535169 @default.
• W2565851976 cites W2048266589 @default.
• W2565851976 cites W2062227835 @default.
• W2565851976 cites W2067523571 @default.
• W2565851976 cites W2095705004 @default.
• W2565851976 cites W2096645269 @default.
• W2565851976 cites W2105102111 @default.
• W2565851976 cites W2108665656 @default.
• W2565851976 cites W2112796928 @default.
• W2565851976 cites W2113606819 @default.
• W2565851976 cites W2116267755 @default.
• W2565851976 cites W2117810994 @default.
• W2565851976 cites W2125085157 @default.
• W2565851976 cites W2125203716 @default.
• W2565851976 cites W2133257461 @default.
• W2565851976 cites W2136189984 @default.
• W2565851976 cites W2136262589 @default.
• W2565851976 cites W2140676977 @default.
• W2565851976 cites W2145285494 @default.
• W2565851976 cites W2145889472 @default.
• W2565851976 cites W2152839228 @default.
• W2565851976 cites W2153846939 @default.
• W2565851976 cites W2155893237 @default.
• W2565851976 cites W2160428323 @default.
• W2565851976 cites W2162390675 @default.
• W2565851976 cites W2163605009 @default.
• W2565851976 cites W2172174689 @default.
• W2565851976 cites W2187230075 @default.
• W2565851976 cites W2252268321 @default.
• W2565851976 cites W2253807446 @default.
• W2565851976 cites W2285660444 @default.
• W2565851976 cites W2546302380 @default.
• W2565851976 cites W2962835968 @default.
• W2565851976 cites W2963674932 @default.
• W2565851976 cites W3151762208 @default.
• W2565851976 cites W1785235667 @default.
• W2565851976 doi "https://doi.org/10.5555/3195638.3195662" @default.
• W2565851976 hasPublicationYear "2016" @default.
• W2565851976 type Work @default.
• W2565851976 sameAs 2565851976 @default.
• W2565851976 citedByCount "198" @default.
• W2565851976 countsByYear W25658519762016 @default.
• W2565851976 countsByYear W25658519762017 @default.
• W2565851976 countsByYear W25658519762018 @default.
• W2565851976 countsByYear W25658519762019 @default.
• W2565851976 countsByYear W25658519762020 @default.
• W2565851976 countsByYear W25658519762021 @default.
• W2565851976 countsByYear W25658519762023 @default.
• W2565851976 crossrefType "proceedings-article" @default.
• W2565851976 hasAuthorship W2565851976A5014715855 @default.
• W2565851976 hasAuthorship W2565851976A5033237869 @default.
• W2565851976 hasAuthorship W2565851976A5034758732 @default.
• W2565851976 hasAuthorship W2565851976A5036492388 @default.
• W2565851976 hasAuthorship W2565851976A5056949827 @default.
• W2565851976 hasAuthorship W2565851976A5063724348 @default.
• W2565851976 hasAuthorship W2565851976A5072861069 @default.
• W2565851976 hasAuthorship W2565851976A5081598354 @default.
• W2565851976 hasConcept C11413529 @default.
• W2565851976 hasConcept C13164978 @default.
• W2565851976 hasConcept C151319957 @default.
• W2565851976 hasConcept C154945302 @default.
• W2565851976 hasConcept C165696696 @default.
• W2565851976 hasConcept C173608175 @default.
• W2565851976 hasConcept C2776257435 @default.
• W2565851976 hasConcept C38652104 @default.
• W2565851976 hasConcept C41008148 @default.
• W2565851976 hasConcept C42935608 @default.
• W2565851976 hasConcept C45374587 @default.
• W2565851976 hasConcept C50644808 @default.
• W2565851976 hasConcept C68339613 @default.
• W2565851976 hasConcept C75165309 @default.
• W2565851976 hasConcept C76155785 @default.
• W2565851976 hasConcept C9390403 @default.
• W2565851976 hasConceptScore W2565851976C11413529 @default.
• W2565851976 hasConceptScore W2565851976C13164978 @default.
• W2565851976 hasConceptScore W2565851976C151319957 @default.

• next