Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287704966> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4287704966 abstract "The design of systems implementing low precision neural networks with emerging memories such as resistive random access memory (RRAM) is a significant lead for reducing the energy consumption of artificial intelligence. To achieve maximum energy efficiency in such systems, logic and memory should be integrated as tightly as possible. In this work, we focus on the case of ternary neural networks, where synaptic weights assume ternary values. We propose a two-transistor/two-resistor memory architecture employing a precharge sense amplifier, where the weight value can be extracted in a single sense operation. Based on experimental measurements on a hybrid 130 nm CMOS/RRAM chip featuring this sense amplifier, we show that this technique is particularly appropriate at low supply voltage, and that it is resilient to process, voltage, and temperature variations. We characterize the bit error rate in our scheme. We show based on neural network simulation on the CIFAR-10 image recognition task that the use of ternary neural networks significantly increases neural network performance, with regards to binary ones, which are often preferred for inference hardware. We finally evidence that the neural network is immune to the type of bit errors observed in our scheme, which can therefore be used without error correction." @default.
- W4287704966 created "2022-07-26" @default.
- W4287704966 creator A5004399656 @default.
- W4287704966 creator A5032162752 @default.
- W4287704966 creator A5046458712 @default.
- W4287704966 creator A5046559880 @default.
- W4287704966 creator A5063819347 @default.
- W4287704966 creator A5077172922 @default.
- W4287704966 creator A5087494328 @default.
- W4287704966 creator A5017966514 @default.
- W4287704966 date "2020-07-26" @default.
- W4287704966 modified "2023-09-29" @default.
- W4287704966 title "Implementation of Ternary Weights with Resistive RAM Using a Single Sense Operation per Synapse" @default.
- W4287704966 doi "https://doi.org/10.48550/arxiv.2007.14234" @default.
- W4287704966 hasPublicationYear "2020" @default.
- W4287704966 type Work @default.
- W4287704966 citedByCount "0" @default.
- W4287704966 crossrefType "posted-content" @default.
- W4287704966 hasAuthorship W4287704966A5004399656 @default.
- W4287704966 hasAuthorship W4287704966A5017966514 @default.
- W4287704966 hasAuthorship W4287704966A5032162752 @default.
- W4287704966 hasAuthorship W4287704966A5046458712 @default.
- W4287704966 hasAuthorship W4287704966A5046559880 @default.
- W4287704966 hasAuthorship W4287704966A5063819347 @default.
- W4287704966 hasAuthorship W4287704966A5077172922 @default.
- W4287704966 hasAuthorship W4287704966A5087494328 @default.
- W4287704966 hasBestOaLocation W42877049661 @default.
- W4287704966 hasConcept C113775141 @default.
- W4287704966 hasConcept C11413529 @default.
- W4287704966 hasConcept C119599485 @default.
- W4287704966 hasConcept C127413603 @default.
- W4287704966 hasConcept C137488568 @default.
- W4287704966 hasConcept C143141573 @default.
- W4287704966 hasConcept C154945302 @default.
- W4287704966 hasConcept C165801399 @default.
- W4287704966 hasConcept C172385210 @default.
- W4287704966 hasConcept C182019814 @default.
- W4287704966 hasConcept C199360897 @default.
- W4287704966 hasConcept C24326235 @default.
- W4287704966 hasConcept C32666082 @default.
- W4287704966 hasConcept C40969351 @default.
- W4287704966 hasConcept C41008148 @default.
- W4287704966 hasConcept C50644808 @default.
- W4287704966 hasConcept C64452783 @default.
- W4287704966 hasConcept C9390403 @default.
- W4287704966 hasConcept C98986596 @default.
- W4287704966 hasConceptScore W4287704966C113775141 @default.
- W4287704966 hasConceptScore W4287704966C11413529 @default.
- W4287704966 hasConceptScore W4287704966C119599485 @default.
- W4287704966 hasConceptScore W4287704966C127413603 @default.
- W4287704966 hasConceptScore W4287704966C137488568 @default.
- W4287704966 hasConceptScore W4287704966C143141573 @default.
- W4287704966 hasConceptScore W4287704966C154945302 @default.
- W4287704966 hasConceptScore W4287704966C165801399 @default.
- W4287704966 hasConceptScore W4287704966C172385210 @default.
- W4287704966 hasConceptScore W4287704966C182019814 @default.
- W4287704966 hasConceptScore W4287704966C199360897 @default.
- W4287704966 hasConceptScore W4287704966C24326235 @default.
- W4287704966 hasConceptScore W4287704966C32666082 @default.
- W4287704966 hasConceptScore W4287704966C40969351 @default.
- W4287704966 hasConceptScore W4287704966C41008148 @default.
- W4287704966 hasConceptScore W4287704966C50644808 @default.
- W4287704966 hasConceptScore W4287704966C64452783 @default.
- W4287704966 hasConceptScore W4287704966C9390403 @default.
- W4287704966 hasConceptScore W4287704966C98986596 @default.
- W4287704966 hasLocation W42877049661 @default.
- W4287704966 hasOpenAccess W4287704966 @default.
- W4287704966 hasPrimaryLocation W42877049661 @default.
- W4287704966 hasRelatedWork W1949239239 @default.
- W4287704966 hasRelatedWork W2060253890 @default.
- W4287704966 hasRelatedWork W2097965268 @default.
- W4287704966 hasRelatedWork W2146647189 @default.
- W4287704966 hasRelatedWork W2218822284 @default.
- W4287704966 hasRelatedWork W2808199063 @default.
- W4287704966 hasRelatedWork W2896189158 @default.
- W4287704966 hasRelatedWork W3045594948 @default.
- W4287704966 hasRelatedWork W3094874291 @default.
- W4287704966 hasRelatedWork W4287704966 @default.
- W4287704966 isParatext "false" @default.
- W4287704966 isRetracted "false" @default.
- W4287704966 workType "article" @default.