FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285292207> ?p ?o ?g. }

• W4285292207 endingPage "49166" @default.
• W4285292207 startingPage "49154" @default.
• W4285292207 abstract "Increasing the energy efficiency of deep learning systems is critical for improving the cognitive capability of edge devices, often battery operated, as well as for data centers, constrained by the total power envelope. Specialized architectures accelerated by analog vector-matrix multipliers (VMMs) can reduce by orders of magnitude the energy per operation, since the reduced precision of analog computation does not undermine the classification accuracy of the neural network. We show an analog vector-matrix multiplier fabricated with industry-standard 0.18 μm CMOS process, exploiting a single-transistor non-volatile analog memory cell and dedicated technology circuit co-design. The design is focused on implementation in neural networks performing offline training. The VMM performs the analog multiplication of a vector of inputs, encoded in the duration of time pulses, times a matrix of weights, encoded in the programmable currents of the memory cells. A 1.72 μm2 memory cell is realized with a single transistor with floating gate, which can be operated as a two-terminal analog memristive device with more than 64 programmable current levels and high Ihigh/Ilow ratio (> 103), tuned by the charge injected in the floating gate. A small-area charge amplifier is used to convert the multiply and accumulate operation result into a voltage. System-level projections based on our measurements and simulations provide a throughput of 333.17 GOps/s and an energy efficiency of 122.3 TOps/J, higher than comparable-precision VMMs reported in the literature, and an equivalent area per cell down to 2.15 μm2, lower than any similar state-of-the-art solution. Of critical importance in view of translation to industry, our proposal uses in a new way an industry-standard low-cost single-poly CMOS process flow." @default.
• W4285292207 created "2022-07-14" @default.
• W4285292207 creator A5020430535 @default.
• W4285292207 creator A5076267890 @default.
• W4285292207 creator A5088596296 @default.
• W4285292207 date "2022-01-01" @default.
• W4285292207 modified "2023-10-14" @default.
• W4285292207 title "Time Domain Analog Neuromorphic Engine Based on High-Density Non-Volatile Memory in Single-Poly CMOS" @default.
• W4285292207 cites W1549382382 @default.
• W4285292207 cites W1560726208 @default.
• W4285292207 cites W1592112976 @default.
• W4285292207 cites W1798990620 @default.
• W4285292207 cites W1964276494 @default.
• W4285292207 cites W1984750698 @default.
• W4285292207 cites W1987727343 @default.
• W4285292207 cites W2014307680 @default.
• W4285292207 cites W2016045400 @default.
• W4285292207 cites W2020769149 @default.
• W4285292207 cites W2044228950 @default.
• W4285292207 cites W2068115049 @default.
• W4285292207 cites W2070731140 @default.
• W4285292207 cites W2132383505 @default.
• W4285292207 cites W2135210684 @default.
• W4285292207 cites W2158284457 @default.
• W4285292207 cites W2309589607 @default.
• W4285292207 cites W2487018225 @default.
• W4285292207 cites W2740220207 @default.
• W4285292207 cites W2752846509 @default.
• W4285292207 cites W2768194408 @default.
• W4285292207 cites W2778345336 @default.
• W4285292207 cites W2787453651 @default.
• W4285292207 cites W2796625795 @default.
• W4285292207 cites W2797406914 @default.
• W4285292207 cites W2803163155 @default.
• W4285292207 cites W2895752922 @default.
• W4285292207 cites W2896122000 @default.
• W4285292207 cites W2912270587 @default.
• W4285292207 cites W2963326222 @default.
• W4285292207 cites W2964686526 @default.
• W4285292207 cites W2966124301 @default.
• W4285292207 cites W2967391386 @default.
• W4285292207 cites W2970404564 @default.
• W4285292207 cites W2977053823 @default.
• W4285292207 cites W2992817139 @default.
• W4285292207 cites W3011635793 @default.
• W4285292207 cites W3013080934 @default.
• W4285292207 cites W3016048022 @default.
• W4285292207 cites W3043159233 @default.
• W4285292207 cites W3097855118 @default.
• W4285292207 cites W3099377273 @default.
• W4285292207 cites W3115328291 @default.
• W4285292207 cites W3115565033 @default.
• W4285292207 cites W3156580923 @default.
• W4285292207 doi "https://doi.org/10.1109/access.2022.3172488" @default.
• W4285292207 hasPublicationYear "2022" @default.
• W4285292207 type Work @default.
• W4285292207 citedByCount "2" @default.
• W4285292207 countsByYear W42852922072022 @default.
• W4285292207 countsByYear W42852922072023 @default.
• W4285292207 crossrefType "journal-article" @default.
• W4285292207 hasAuthorship W4285292207A5020430535 @default.
• W4285292207 hasAuthorship W4285292207A5076267890 @default.
• W4285292207 hasAuthorship W4285292207A5088596296 @default.
• W4285292207 hasBestOaLocation W42852922071 @default.
• W4285292207 hasConcept C119599485 @default.
• W4285292207 hasConcept C127413603 @default.
• W4285292207 hasConcept C13412647 @default.
• W4285292207 hasConcept C134146338 @default.
• W4285292207 hasConcept C149128552 @default.
• W4285292207 hasConcept C151927369 @default.
• W4285292207 hasConcept C154945302 @default.
• W4285292207 hasConcept C165801399 @default.
• W4285292207 hasConcept C172385210 @default.
• W4285292207 hasConcept C24326235 @default.
• W4285292207 hasConcept C29074008 @default.
• W4285292207 hasConcept C41008148 @default.
• W4285292207 hasConcept C46362747 @default.
• W4285292207 hasConcept C50644808 @default.
• W4285292207 hasConcept C84462506 @default.
• W4285292207 hasConcept C9390403 @default.
• W4285292207 hasConcept C98142538 @default.
• W4285292207 hasConceptScore W4285292207C119599485 @default.
• W4285292207 hasConceptScore W4285292207C127413603 @default.
• W4285292207 hasConceptScore W4285292207C13412647 @default.
• W4285292207 hasConceptScore W4285292207C134146338 @default.
• W4285292207 hasConceptScore W4285292207C149128552 @default.
• W4285292207 hasConceptScore W4285292207C151927369 @default.
• W4285292207 hasConceptScore W4285292207C154945302 @default.
• W4285292207 hasConceptScore W4285292207C165801399 @default.
• W4285292207 hasConceptScore W4285292207C172385210 @default.
• W4285292207 hasConceptScore W4285292207C24326235 @default.
• W4285292207 hasConceptScore W4285292207C29074008 @default.
• W4285292207 hasConceptScore W4285292207C41008148 @default.
• W4285292207 hasConceptScore W4285292207C46362747 @default.
• W4285292207 hasConceptScore W4285292207C50644808 @default.
• W4285292207 hasConceptScore W4285292207C84462506 @default.
• W4285292207 hasConceptScore W4285292207C9390403 @default.
• W4285292207 hasConceptScore W4285292207C98142538 @default.

• next