Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385080007> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4385080007 abstract "This paper proposes an efficient hardware accelerator named EACNN for use in Convolution Neural Networks. EACNN is an efficient CNN architecture that is based on co-optimization of algorithms and hardware. The proposed approach is based on linear approximation of the weights for pre-trained networks with low loss of accuracy. Furthermore, a weight substitution and remapping technique adopts linear approximation coefficients to replace CNN weights. That leads to a repetition of the weight values across different kernels and enables the reuse of CNN computations for various output feature maps. The input activations corresponding to the same linear co-efficient can be multiplied and accumulated first and then reused to generate multiple output feature maps. This computational reuse method reduces the number of multiplication and addition operations and memory accesses, which is efficiently supported by a dedicated element in the proposed EACNN. Experimental results on CIFAR 10 and CIFAR 100 datasets show that the proposed method eliminates around 61% of the multiplications in the network without significant loss of accuracy <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$(< 3%)$</tex> . As a demonstration, a hardware accelerator based on EACNN was implemented on Xilinx FPGA Artix 7 and achieved a 50% reduction in the FPGA hardware resources." @default.
- W4385080007 created "2023-07-23" @default.
- W4385080007 creator A5013814572 @default.
- W4385080007 creator A5020682177 @default.
- W4385080007 creator A5045307151 @default.
- W4385080007 creator A5078108496 @default.
- W4385080007 creator A5091151249 @default.
- W4385080007 date "2023-05-21" @default.
- W4385080007 modified "2023-09-24" @default.
- W4385080007 title "EACNN: Efficient CNN Accelerator Utilizing Linear Approximation and Computation Reuse" @default.
- W4385080007 cites W2147768505 @default.
- W4385080007 cites W2194775991 @default.
- W4385080007 cites W2289252105 @default.
- W4385080007 cites W2612076670 @default.
- W4385080007 cites W2748818695 @default.
- W4385080007 cites W2798170643 @default.
- W4385080007 cites W2906663849 @default.
- W4385080007 cites W3161926058 @default.
- W4385080007 cites W3205311052 @default.
- W4385080007 doi "https://doi.org/10.1109/iscas46773.2023.10181343" @default.
- W4385080007 hasPublicationYear "2023" @default.
- W4385080007 type Work @default.
- W4385080007 citedByCount "0" @default.
- W4385080007 crossrefType "proceedings-article" @default.
- W4385080007 hasAuthorship W4385080007A5013814572 @default.
- W4385080007 hasAuthorship W4385080007A5020682177 @default.
- W4385080007 hasAuthorship W4385080007A5045307151 @default.
- W4385080007 hasAuthorship W4385080007A5078108496 @default.
- W4385080007 hasAuthorship W4385080007A5091151249 @default.
- W4385080007 hasConcept C113775141 @default.
- W4385080007 hasConcept C11413529 @default.
- W4385080007 hasConcept C13164978 @default.
- W4385080007 hasConcept C138885662 @default.
- W4385080007 hasConcept C154945302 @default.
- W4385080007 hasConcept C173608175 @default.
- W4385080007 hasConcept C18903297 @default.
- W4385080007 hasConcept C199360897 @default.
- W4385080007 hasConcept C206588197 @default.
- W4385080007 hasConcept C2776401178 @default.
- W4385080007 hasConcept C2777904410 @default.
- W4385080007 hasConcept C41008148 @default.
- W4385080007 hasConcept C41895202 @default.
- W4385080007 hasConcept C42935608 @default.
- W4385080007 hasConcept C45347329 @default.
- W4385080007 hasConcept C45374587 @default.
- W4385080007 hasConcept C459310 @default.
- W4385080007 hasConcept C50644808 @default.
- W4385080007 hasConcept C65232700 @default.
- W4385080007 hasConcept C81363708 @default.
- W4385080007 hasConcept C86803240 @default.
- W4385080007 hasConcept C9390403 @default.
- W4385080007 hasConceptScore W4385080007C113775141 @default.
- W4385080007 hasConceptScore W4385080007C11413529 @default.
- W4385080007 hasConceptScore W4385080007C13164978 @default.
- W4385080007 hasConceptScore W4385080007C138885662 @default.
- W4385080007 hasConceptScore W4385080007C154945302 @default.
- W4385080007 hasConceptScore W4385080007C173608175 @default.
- W4385080007 hasConceptScore W4385080007C18903297 @default.
- W4385080007 hasConceptScore W4385080007C199360897 @default.
- W4385080007 hasConceptScore W4385080007C206588197 @default.
- W4385080007 hasConceptScore W4385080007C2776401178 @default.
- W4385080007 hasConceptScore W4385080007C2777904410 @default.
- W4385080007 hasConceptScore W4385080007C41008148 @default.
- W4385080007 hasConceptScore W4385080007C41895202 @default.
- W4385080007 hasConceptScore W4385080007C42935608 @default.
- W4385080007 hasConceptScore W4385080007C45347329 @default.
- W4385080007 hasConceptScore W4385080007C45374587 @default.
- W4385080007 hasConceptScore W4385080007C459310 @default.
- W4385080007 hasConceptScore W4385080007C50644808 @default.
- W4385080007 hasConceptScore W4385080007C65232700 @default.
- W4385080007 hasConceptScore W4385080007C81363708 @default.
- W4385080007 hasConceptScore W4385080007C86803240 @default.
- W4385080007 hasConceptScore W4385080007C9390403 @default.
- W4385080007 hasLocation W43850800071 @default.
- W4385080007 hasOpenAccess W4385080007 @default.
- W4385080007 hasPrimaryLocation W43850800071 @default.
- W4385080007 hasRelatedWork W2295680811 @default.
- W4385080007 hasRelatedWork W2561485601 @default.
- W4385080007 hasRelatedWork W2565158842 @default.
- W4385080007 hasRelatedWork W2770717529 @default.
- W4385080007 hasRelatedWork W2949649078 @default.
- W4385080007 hasRelatedWork W2952735458 @default.
- W4385080007 hasRelatedWork W4308067493 @default.
- W4385080007 hasRelatedWork W4319662858 @default.
- W4385080007 hasRelatedWork W4327640786 @default.
- W4385080007 hasRelatedWork W3116497574 @default.
- W4385080007 isParatext "false" @default.
- W4385080007 isRetracted "false" @default.
- W4385080007 workType "article" @default.