Matches in SemOpenAlex for { <https://semopenalex.org/work/W3157383824> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W3157383824 abstract "Binary Neural Networks enable smart IoT devices, as they significantly reduce the required memory footprint and computational complexity while retaining a high network performance and flexibility. This paper presents ChewBaccaNN, a 0.7 mm <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>2</sup> sized binary convolutional neural network (CNN) accelerator designed in GlobalFoundries 22 nm technology. By exploiting efficient data re-use, data buffering, latch-based memories, and voltage scaling, a throughput of 241 GOPS is achieved while consuming just 1.1 mW at 0.4V/154MHz during inference of binary CNNs with up to 7×7 kernels, leading to a peak core energy efficiency of 223 TOPS/W. ChewBaccaNN's flexibility allows to run a much wider range of binary CNNs than other accelerators, drastically improving the accuracy-energy tradeoff beyond what can be captured by the TOPS/W metric. In fact, it can perform CIFAR-10 inference at 86.8% accuracy with merely 1.3 J, thus exceeding the accuracy while at the same time lowering the energy cost by 2.8× compared to even the most efficient and much larger analog processing-in-memory devices, while keeping the flexibility of running larger CNNs for higher accuracy when needed. It also runs a binary ResNet-18 trained on the 1000-class ILSVRC dataset and improves the energy efficiency by 4.4× over accelerators of similar flexibility. Furthermore, it can perform inference on a binarized ResNet-18 trained with 8-bases Group-Net to achieve a 67.5% Top-1 accuracy with only 3.0mJ/frame-at an accuracy drop of merely 1.8% from the fullprecision ResNet-18." @default.
- W3157383824 created "2021-05-10" @default.
- W3157383824 creator A5025399641 @default.
- W3157383824 creator A5043163538 @default.
- W3157383824 creator A5043408422 @default.
- W3157383824 creator A5050957615 @default.
- W3157383824 date "2021-05-01" @default.
- W3157383824 modified "2023-10-11" @default.
- W3157383824 title "ChewBaccaNN: A Flexible 223 TOPS/W BNN Accelerator" @default.
- W3157383824 cites W2604319603 @default.
- W3157383824 cites W2604814848 @default.
- W3157383824 cites W2845210056 @default.
- W3157383824 cites W2884821289 @default.
- W3157383824 cites W2894696827 @default.
- W3157383824 cites W2895531329 @default.
- W3157383824 cites W2920326572 @default.
- W3157383824 cites W2963122961 @default.
- W3157383824 cites W2963221280 @default.
- W3157383824 cites W2963255460 @default.
- W3157383824 cites W2963258090 @default.
- W3157383824 cites W2964195914 @default.
- W3157383824 cites W2964242827 @default.
- W3157383824 cites W2982317364 @default.
- W3157383824 cites W3008515144 @default.
- W3157383824 cites W3048617363 @default.
- W3157383824 cites W3119780625 @default.
- W3157383824 doi "https://doi.org/10.1109/iscas51556.2021.9401214" @default.
- W3157383824 hasPublicationYear "2021" @default.
- W3157383824 type Work @default.
- W3157383824 sameAs 3157383824 @default.
- W3157383824 citedByCount "6" @default.
- W3157383824 countsByYear W31573838242021 @default.
- W3157383824 countsByYear W31573838242022 @default.
- W3157383824 countsByYear W31573838242023 @default.
- W3157383824 crossrefType "proceedings-article" @default.
- W3157383824 hasAuthorship W3157383824A5025399641 @default.
- W3157383824 hasAuthorship W3157383824A5043163538 @default.
- W3157383824 hasAuthorship W3157383824A5043408422 @default.
- W3157383824 hasAuthorship W3157383824A5050957615 @default.
- W3157383824 hasBestOaLocation W31573838242 @default.
- W3157383824 hasConcept C105795698 @default.
- W3157383824 hasConcept C111919701 @default.
- W3157383824 hasConcept C113775141 @default.
- W3157383824 hasConcept C11413529 @default.
- W3157383824 hasConcept C154945302 @default.
- W3157383824 hasConcept C157764524 @default.
- W3157383824 hasConcept C159737794 @default.
- W3157383824 hasConcept C173608175 @default.
- W3157383824 hasConcept C186370098 @default.
- W3157383824 hasConcept C2524010 @default.
- W3157383824 hasConcept C2776214188 @default.
- W3157383824 hasConcept C2777675136 @default.
- W3157383824 hasConcept C2780598303 @default.
- W3157383824 hasConcept C33923547 @default.
- W3157383824 hasConcept C41008148 @default.
- W3157383824 hasConcept C48372109 @default.
- W3157383824 hasConcept C555944384 @default.
- W3157383824 hasConcept C74912251 @default.
- W3157383824 hasConcept C81363708 @default.
- W3157383824 hasConcept C94375191 @default.
- W3157383824 hasConceptScore W3157383824C105795698 @default.
- W3157383824 hasConceptScore W3157383824C111919701 @default.
- W3157383824 hasConceptScore W3157383824C113775141 @default.
- W3157383824 hasConceptScore W3157383824C11413529 @default.
- W3157383824 hasConceptScore W3157383824C154945302 @default.
- W3157383824 hasConceptScore W3157383824C157764524 @default.
- W3157383824 hasConceptScore W3157383824C159737794 @default.
- W3157383824 hasConceptScore W3157383824C173608175 @default.
- W3157383824 hasConceptScore W3157383824C186370098 @default.
- W3157383824 hasConceptScore W3157383824C2524010 @default.
- W3157383824 hasConceptScore W3157383824C2776214188 @default.
- W3157383824 hasConceptScore W3157383824C2777675136 @default.
- W3157383824 hasConceptScore W3157383824C2780598303 @default.
- W3157383824 hasConceptScore W3157383824C33923547 @default.
- W3157383824 hasConceptScore W3157383824C41008148 @default.
- W3157383824 hasConceptScore W3157383824C48372109 @default.
- W3157383824 hasConceptScore W3157383824C555944384 @default.
- W3157383824 hasConceptScore W3157383824C74912251 @default.
- W3157383824 hasConceptScore W3157383824C81363708 @default.
- W3157383824 hasConceptScore W3157383824C94375191 @default.
- W3157383824 hasLocation W31573838241 @default.
- W3157383824 hasLocation W31573838242 @default.
- W3157383824 hasOpenAccess W3157383824 @default.
- W3157383824 hasPrimaryLocation W31573838241 @default.
- W3157383824 hasRelatedWork W2047588290 @default.
- W3157383824 hasRelatedWork W2466675884 @default.
- W3157383824 hasRelatedWork W2524802307 @default.
- W3157383824 hasRelatedWork W2581650043 @default.
- W3157383824 hasRelatedWork W2748428003 @default.
- W3157383824 hasRelatedWork W2791431468 @default.
- W3157383824 hasRelatedWork W2995097845 @default.
- W3157383824 hasRelatedWork W2997828269 @default.
- W3157383824 hasRelatedWork W4287991338 @default.
- W3157383824 hasRelatedWork W4289729660 @default.
- W3157383824 isParatext "false" @default.
- W3157383824 isRetracted "false" @default.
- W3157383824 magId "3157383824" @default.
- W3157383824 workType "article" @default.