Matches in SemOpenAlex for { <https://semopenalex.org/work/W2783501677> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W2783501677 abstract "Resistive memory crossbars can dramatically reduce the energy required to perform computations in neural algorithms by three orders of magnitude when compared to an optimized digital ASIC [1]. For data intensive applications, the computational energy is dominated by moving data between the processor, SRAM, and DRAM. Analog crossbars overcome this by allowing data to be processed directly at each memory element. Analog crossbars accelerate three key operations that are the bulk of the computation in a neural network as illustrated in Fig 1: vector matrix multiplies (VMM), matrix vector multiplies (MVM), and outer product rank 1 updates (OPU)[2]. For an NxN crossbar the energy for each operation scales as the number of memory elements O(N2) [2]. This is because the crossbar performs its entire computation in one step, charging all the capacitances only once. Thus the CV2 energy of the array scales as array size. This fundamentally better than trying to read or write a digital memory. Each row of any NxN digital memory must be accessed one at a time, resulting in N columns of length O(N) being charged N times, requiring O(N3) energy to read a digital memory. Thus an analog crossbar has a fundamental O(N) energy scaling advantage over a digital system. Furthermore, if the read operation is done at low voltage and is therefore noise limited, the read energy can even be independent of the crossbar size, O(1) [2]." @default.
- W2783501677 created "2018-01-26" @default.
- W2783501677 creator A5001046383 @default.
- W2783501677 creator A5010408777 @default.
- W2783501677 creator A5017426058 @default.
- W2783501677 creator A5056067427 @default.
- W2783501677 creator A5063028481 @default.
- W2783501677 creator A5068520992 @default.
- W2783501677 creator A5085659212 @default.
- W2783501677 date "2017-10-01" @default.
- W2783501677 modified "2023-10-14" @default.
- W2783501677 title "Designing an analog crossbar based neuromorphic accelerator" @default.
- W2783501677 cites W2282030913 @default.
- W2783501677 cites W2552299751 @default.
- W2783501677 cites W2554279936 @default.
- W2783501677 cites W2591029953 @default.
- W2783501677 cites W2743575928 @default.
- W2783501677 cites W2772823933 @default.
- W2783501677 doi "https://doi.org/10.1109/e3s.2017.8246155" @default.
- W2783501677 hasPublicationYear "2017" @default.
- W2783501677 type Work @default.
- W2783501677 sameAs 2783501677 @default.
- W2783501677 citedByCount "4" @default.
- W2783501677 countsByYear W27835016772018 @default.
- W2783501677 countsByYear W27835016772020 @default.
- W2783501677 countsByYear W27835016772021 @default.
- W2783501677 countsByYear W27835016772022 @default.
- W2783501677 crossrefType "proceedings-article" @default.
- W2783501677 hasAuthorship W2783501677A5001046383 @default.
- W2783501677 hasAuthorship W2783501677A5010408777 @default.
- W2783501677 hasAuthorship W2783501677A5017426058 @default.
- W2783501677 hasAuthorship W2783501677A5056067427 @default.
- W2783501677 hasAuthorship W2783501677A5063028481 @default.
- W2783501677 hasAuthorship W2783501677A5068520992 @default.
- W2783501677 hasAuthorship W2783501677A5085659212 @default.
- W2783501677 hasBestOaLocation W27835016772 @default.
- W2783501677 hasConcept C118524514 @default.
- W2783501677 hasConcept C151927369 @default.
- W2783501677 hasConcept C154945302 @default.
- W2783501677 hasConcept C29984679 @default.
- W2783501677 hasConcept C41008148 @default.
- W2783501677 hasConcept C50644808 @default.
- W2783501677 hasConcept C76155785 @default.
- W2783501677 hasConceptScore W2783501677C118524514 @default.
- W2783501677 hasConceptScore W2783501677C151927369 @default.
- W2783501677 hasConceptScore W2783501677C154945302 @default.
- W2783501677 hasConceptScore W2783501677C29984679 @default.
- W2783501677 hasConceptScore W2783501677C41008148 @default.
- W2783501677 hasConceptScore W2783501677C50644808 @default.
- W2783501677 hasConceptScore W2783501677C76155785 @default.
- W2783501677 hasLocation W27835016771 @default.
- W2783501677 hasLocation W27835016772 @default.
- W2783501677 hasLocation W27835016773 @default.
- W2783501677 hasOpenAccess W2783501677 @default.
- W2783501677 hasPrimaryLocation W27835016771 @default.
- W2783501677 hasRelatedWork W1973433558 @default.
- W2783501677 hasRelatedWork W2001480795 @default.
- W2783501677 hasRelatedWork W2075493167 @default.
- W2783501677 hasRelatedWork W2620982695 @default.
- W2783501677 hasRelatedWork W2807081766 @default.
- W2783501677 hasRelatedWork W2989604810 @default.
- W2783501677 hasRelatedWork W2991631906 @default.
- W2783501677 hasRelatedWork W3025685192 @default.
- W2783501677 hasRelatedWork W3092023946 @default.
- W2783501677 hasRelatedWork W4308098692 @default.
- W2783501677 isParatext "false" @default.
- W2783501677 isRetracted "false" @default.
- W2783501677 magId "2783501677" @default.
- W2783501677 workType "article" @default.