Matches in SemOpenAlex for { <https://semopenalex.org/work/W2808976322> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2808976322 endingPage "2731" @default.
- W2808976322 startingPage "2722" @default.
- W2808976322 abstract "This paper presents a 28-nm system-on-chip (SoC) for Internet of things (IoT) applications with a programmable accelerator design that implements a powerful fully connected deep neural network (DNN) classifier. To reach the required low energy consumption, we exploit the key properties of neural network algorithms: parallelism, data reuse, small/sparse data, and noise tolerance. We map the algorithm to a very large scale integration (VLSI) architecture based around an singleinstruction, multiple-data data path with hardware support to exploit data sparsity by completely eliding unnecessary computation and data movement. This approach exploits sparsity, without compromising the parallel computation. We also exploit the inherent algorithmic noise-tolerance of neural networks, by introducing circuit-level timing violation detection to allow worst case voltage guard-bands to be minimized. The resulting intermittent timing violations may result in logic errors, which conventionally need to be corrected. However, in lieu of explicit error correction, we cope with this by accentuating the noise tolerance of neural networks. The measured test chip achieves high classification accuracy (98.36% for the MNIST test set), while tolerating aggregate timing violation rates>10 <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>-1</sup> . The accelerator achieves a minimum energy of 0.36 μJ/inference at 667 MHz; maximum throughput at 1.2 GHz and 0.57 μJ/inference; or a 10% margined operating point at 1 GHz and 0.58 μJ/inference." @default.
- W2808976322 created "2018-06-29" @default.
- W2808976322 creator A5026496503 @default.
- W2808976322 creator A5043327132 @default.
- W2808976322 creator A5062046480 @default.
- W2808976322 creator A5088853105 @default.
- W2808976322 date "2018-09-01" @default.
- W2808976322 modified "2023-10-10" @default.
- W2808976322 title "DNN Engine: A 28-nm Timing-Error Tolerant Sparse Deep Neural Network Processor for IoT Applications" @default.
- W2808976322 cites W2026359221 @default.
- W2808976322 cites W2057558141 @default.
- W2808976322 cites W2076344933 @default.
- W2808976322 cites W2100215874 @default.
- W2808976322 cites W2138913040 @default.
- W2808976322 cites W2141762573 @default.
- W2808976322 cites W2290132443 @default.
- W2808976322 cites W2513554817 @default.
- W2808976322 cites W2591601611 @default.
- W2808976322 cites W2593197369 @default.
- W2808976322 cites W2594492285 @default.
- W2808976322 cites W2595408789 @default.
- W2808976322 cites W2745584882 @default.
- W2808976322 cites W2761878354 @default.
- W2808976322 cites W2770934218 @default.
- W2808976322 cites W2777372517 @default.
- W2808976322 cites W2792893539 @default.
- W2808976322 cites W2963029056 @default.
- W2808976322 cites W2963893493 @default.
- W2808976322 cites W3024621361 @default.
- W2808976322 cites W3141370589 @default.
- W2808976322 cites W4236432903 @default.
- W2808976322 cites W4252174618 @default.
- W2808976322 doi "https://doi.org/10.1109/jssc.2018.2841824" @default.
- W2808976322 hasPublicationYear "2018" @default.
- W2808976322 type Work @default.
- W2808976322 sameAs 2808976322 @default.
- W2808976322 citedByCount "73" @default.
- W2808976322 countsByYear W28089763222019 @default.
- W2808976322 countsByYear W28089763222020 @default.
- W2808976322 countsByYear W28089763222021 @default.
- W2808976322 countsByYear W28089763222022 @default.
- W2808976322 countsByYear W28089763222023 @default.
- W2808976322 crossrefType "journal-article" @default.
- W2808976322 hasAuthorship W2808976322A5026496503 @default.
- W2808976322 hasAuthorship W2808976322A5043327132 @default.
- W2808976322 hasAuthorship W2808976322A5062046480 @default.
- W2808976322 hasAuthorship W2808976322A5088853105 @default.
- W2808976322 hasConcept C111919701 @default.
- W2808976322 hasConcept C113775141 @default.
- W2808976322 hasConcept C14580979 @default.
- W2808976322 hasConcept C149635348 @default.
- W2808976322 hasConcept C154945302 @default.
- W2808976322 hasConcept C165696696 @default.
- W2808976322 hasConcept C173608175 @default.
- W2808976322 hasConcept C190502265 @default.
- W2808976322 hasConcept C2776214188 @default.
- W2808976322 hasConcept C2779960059 @default.
- W2808976322 hasConcept C38652104 @default.
- W2808976322 hasConcept C41008148 @default.
- W2808976322 hasConcept C50644808 @default.
- W2808976322 hasConceptScore W2808976322C111919701 @default.
- W2808976322 hasConceptScore W2808976322C113775141 @default.
- W2808976322 hasConceptScore W2808976322C14580979 @default.
- W2808976322 hasConceptScore W2808976322C149635348 @default.
- W2808976322 hasConceptScore W2808976322C154945302 @default.
- W2808976322 hasConceptScore W2808976322C165696696 @default.
- W2808976322 hasConceptScore W2808976322C173608175 @default.
- W2808976322 hasConceptScore W2808976322C190502265 @default.
- W2808976322 hasConceptScore W2808976322C2776214188 @default.
- W2808976322 hasConceptScore W2808976322C2779960059 @default.
- W2808976322 hasConceptScore W2808976322C38652104 @default.
- W2808976322 hasConceptScore W2808976322C41008148 @default.
- W2808976322 hasConceptScore W2808976322C50644808 @default.
- W2808976322 hasFunder F4320307102 @default.
- W2808976322 hasFunder F4320332180 @default.
- W2808976322 hasIssue "9" @default.
- W2808976322 hasLocation W28089763221 @default.
- W2808976322 hasOpenAccess W2808976322 @default.
- W2808976322 hasPrimaryLocation W28089763221 @default.
- W2808976322 hasRelatedWork W1495791788 @default.
- W2808976322 hasRelatedWork W1990581372 @default.
- W2808976322 hasRelatedWork W2513409985 @default.
- W2808976322 hasRelatedWork W2792016738 @default.
- W2808976322 hasRelatedWork W2966207885 @default.
- W2808976322 hasRelatedWork W2997512100 @default.
- W2808976322 hasRelatedWork W3042265279 @default.
- W2808976322 hasRelatedWork W3093563898 @default.
- W2808976322 hasRelatedWork W4280607397 @default.
- W2808976322 hasRelatedWork W4287636201 @default.
- W2808976322 hasVolume "53" @default.
- W2808976322 isParatext "false" @default.
- W2808976322 isRetracted "false" @default.
- W2808976322 magId "2808976322" @default.
- W2808976322 workType "article" @default.