FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2954464432> ?p ?o ?g. }

• W2954464432 abstract "Low-power potential of mixed-signal design makes it an alluring option to accelerate Deep Neural Networks (DNNs). However, mixed-signal circuitry suffers from limited range for information encoding, susceptibility to noise, and Analog to Digital (A/D) conversion overheads. This paper aims to address these challenges by offering and leveraging the insight that a vector dot-product (the basic operation in DNNs) can be bit-partitioned into groups of spatially parallel low-bitwidth operations, and interleaved across multiple elements of the vectors. As such, the building blocks of our accelerator become a group of wide, yet low-bitwidth multiply-accumulate units that operate in the analog domain and share a single A/D converter. The low-bitwidth operation tackles the encoding range limitation and facilitates noise mitigation. Moreover, we utilize the switched-capacitor design for our bit-level reformulation of DNN operations. The proposed switched-capacitor circuitry performs the group multiplications in the charge domain and accumulates the results of the group in its capacitors over multiple cycles. The capacitive accumulation combined with wide bit-partitioned operations alleviate the need for A/D conversion per operation. With such mathematical reformulation and its switched-capacitor implementation, we define a 3D-stacked microarchitecture, dubbed BIHIWE." @default.
• W2954464432 created "2019-07-12" @default.
• W2954464432 creator A5000799803 @default.
• W2954464432 creator A5025604916 @default.
• W2954464432 creator A5037648751 @default.
• W2954464432 creator A5067337700 @default.
• W2954464432 creator A5070172290 @default.
• W2954464432 creator A5082318034 @default.
• W2954464432 creator A5082499242 @default.
• W2954464432 creator A5084514143 @default.
• W2954464432 date "2019-06-27" @default.
• W2954464432 modified "2023-09-28" @default.
• W2954464432 title "Mixed-Signal Charge-Domain Acceleration of Deep Neural networks through Interleaved Bit-Partitioned Arithmetic" @default.
• W2954464432 cites W1564201208 @default.
• W2954464432 cites W1598866093 @default.
• W2954464432 cites W1632114991 @default.
• W2954464432 cites W1686810756 @default.
• W2954464432 cites W1980856931 @default.
• W2954464432 cites W1981220134 @default.
• W2954464432 cites W2006312753 @default.
• W2954464432 cites W2010958479 @default.
• W2954464432 cites W2048251158 @default.
• W2954464432 cites W2048266589 @default.
• W2954464432 cites W2064675550 @default.
• W2954464432 cites W2065439108 @default.
• W2954464432 cites W2069743665 @default.
• W2954464432 cites W2082359064 @default.
• W2954464432 cites W2094756095 @default.
• W2954464432 cites W2097117768 @default.
• W2954464432 cites W2108598243 @default.
• W2954464432 cites W2133834148 @default.
• W2954464432 cites W2151631707 @default.
• W2954464432 cites W2152839228 @default.
• W2954464432 cites W2153331583 @default.
• W2954464432 cites W2159742193 @default.
• W2954464432 cites W2166250385 @default.
• W2954464432 cites W2187230075 @default.
• W2954464432 cites W2194775991 @default.
• W2954464432 cites W2261808795 @default.
• W2954464432 cites W2285660444 @default.
• W2954464432 cites W2289252105 @default.
• W2954464432 cites W2294038065 @default.
• W2954464432 cites W2405920868 @default.
• W2954464432 cites W2431931973 @default.
• W2954464432 cites W2442974303 @default.
• W2954464432 cites W2469490737 @default.
• W2954464432 cites W2508602506 @default.
• W2954464432 cites W2511743527 @default.
• W2954464432 cites W2516141709 @default.
• W2954464432 cites W2518281301 @default.
• W2954464432 cites W2518511512 @default.
• W2954464432 cites W2562773490 @default.
• W2954464432 cites W2565851976 @default.
• W2954464432 cites W2587415333 @default.
• W2954464432 cites W2605347906 @default.
• W2954464432 cites W2606722458 @default.
• W2954464432 cites W2613989746 @default.
• W2954464432 cites W2625457103 @default.
• W2954464432 cites W2725092576 @default.
• W2954464432 cites W2742044963 @default.
• W2954464432 cites W2793168176 @default.
• W2954464432 cites W2794141774 @default.
• W2954464432 cites W2796347433 @default.
• W2954464432 cites W2799011136 @default.
• W2954464432 cites W2799229073 @default.
• W2954464432 cites W2809171749 @default.
• W2954464432 cites W2811080765 @default.
• W2954464432 cites W2883542588 @default.
• W2954464432 cites W2884159178 @default.
• W2954464432 cites W2885689789 @default.
• W2954464432 cites W2886601525 @default.
• W2954464432 cites W2887784126 @default.
• W2954464432 cites W2899771611 @default.
• W2954464432 cites W2950458216 @default.
• W2954464432 cites W2950894517 @default.
• W2954464432 cites W2951795638 @default.
• W2954464432 cites W2951995763 @default.
• W2954464432 cites W2952857977 @default.
• W2954464432 cites W2963594949 @default.
• W2954464432 cites W2964057022 @default.
• W2954464432 cites W2964194679 @default.
• W2954464432 cites W3118608800 @default.
• W2954464432 cites W3208968887 @default.
• W2954464432 hasPublicationYear "2019" @default.
• W2954464432 type Work @default.
• W2954464432 sameAs 2954464432 @default.
• W2954464432 citedByCount "2" @default.
• W2954464432 countsByYear W29544644322019 @default.
• W2954464432 countsByYear W29544644322021 @default.
• W2954464432 crossrefType "posted-content" @default.
• W2954464432 hasAuthorship W2954464432A5000799803 @default.
• W2954464432 hasAuthorship W2954464432A5025604916 @default.
• W2954464432 hasAuthorship W2954464432A5037648751 @default.
• W2954464432 hasAuthorship W2954464432A5067337700 @default.
• W2954464432 hasAuthorship W2954464432A5070172290 @default.
• W2954464432 hasAuthorship W2954464432A5082318034 @default.
• W2954464432 hasAuthorship W2954464432A5082499242 @default.
• W2954464432 hasAuthorship W2954464432A5084514143 @default.
• W2954464432 hasConcept C103357873 @default.
• W2954464432 hasConcept C115961682 @default.

• next