FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3020222689> ?p ?o ?g. }

• W3020222689 abstract "The design of systems implementing low precision neural networks with emerging memories such as resistive random access memory (RRAM) is a major lead for reducing the energy consumption of artificial intelligence (AI). Multiple works have for example proposed in-memory architectures to implement low power binarized neural networks. These simple neural networks, where synaptic weights and neuronal activations assume binary values, can indeed approach state-of-the-art performance on vision tasks. In this work, we revisit one of these architectures where synapses are implemented in a differential fashion to reduce bit errors, and synaptic weights are read using precharge sense amplifiers. Based on experimental measurements on a hybrid 130 nm CMOS/RRAM chip and on circuit simulation, we show that the same memory array architecture can be used to implement ternary weights instead of binary weights, and that this technique is particularly appropriate if the sense amplifier is operated in near-threshold regime. We also show based on neural network simulation on the CIFAR-10 image recognition task that going from binary to ternary neural networks significantly increases neural network performance. These results highlight that AI circuits function may sometimes be revisited when operated in low power regimes." @default.
• W3020222689 created "2020-05-01" @default.
• W3020222689 creator A5004399656 @default.
• W3020222689 creator A5025654548 @default.
• W3020222689 creator A5027738561 @default.
• W3020222689 creator A5031158647 @default.
• W3020222689 creator A5046559880 @default.
• W3020222689 creator A5062637694 @default.
• W3020222689 creator A5063819347 @default.
• W3020222689 creator A5064212027 @default.
• W3020222689 creator A5077172922 @default.
• W3020222689 date "2020-08-01" @default.
• W3020222689 modified "2023-09-30" @default.
• W3020222689 title "Low Power In-Memory Implementation of Ternary Neural Networks with Resistive RAM-Based Synapse" @default.
• W3020222689 cites W1998525920 @default.
• W3020222689 cites W2110276925 @default.
• W3020222689 cites W2124799219 @default.
• W3020222689 cites W2583383421 @default.
• W3020222689 cites W2585407525 @default.
• W3020222689 cites W2588666075 @default.
• W3020222689 cites W2796625795 @default.
• W3020222689 cites W2897004326 @default.
• W3020222689 cites W2898399894 @default.
• W3020222689 cites W2912358078 @default.
• W3020222689 cites W2913071218 @default.
• W3020222689 cites W2946942415 @default.
• W3020222689 cites W2962939807 @default.
• W3020222689 cites W2963526839 @default.
• W3020222689 doi "https://doi.org/10.1109/aicas48895.2020.9073877" @default.
• W3020222689 hasPublicationYear "2020" @default.
• W3020222689 type Work @default.
• W3020222689 sameAs 3020222689 @default.
• W3020222689 citedByCount "5" @default.
• W3020222689 countsByYear W30202226892020 @default.
• W3020222689 countsByYear W30202226892021 @default.
• W3020222689 countsByYear W30202226892023 @default.
• W3020222689 crossrefType "proceedings-article" @default.
• W3020222689 hasAuthorship W3020222689A5004399656 @default.
• W3020222689 hasAuthorship W3020222689A5025654548 @default.
• W3020222689 hasAuthorship W3020222689A5027738561 @default.
• W3020222689 hasAuthorship W3020222689A5031158647 @default.
• W3020222689 hasAuthorship W3020222689A5046559880 @default.
• W3020222689 hasAuthorship W3020222689A5062637694 @default.
• W3020222689 hasAuthorship W3020222689A5063819347 @default.
• W3020222689 hasAuthorship W3020222689A5064212027 @default.
• W3020222689 hasAuthorship W3020222689A5077172922 @default.
• W3020222689 hasBestOaLocation W30202226892 @default.
• W3020222689 hasConcept C113775141 @default.
• W3020222689 hasConcept C119599485 @default.
• W3020222689 hasConcept C127413603 @default.
• W3020222689 hasConcept C154945302 @default.
• W3020222689 hasConcept C165801399 @default.
• W3020222689 hasConcept C182019814 @default.
• W3020222689 hasConcept C2778618852 @default.
• W3020222689 hasConcept C32666082 @default.
• W3020222689 hasConcept C33923547 @default.
• W3020222689 hasConcept C41008148 @default.
• W3020222689 hasConcept C48372109 @default.
• W3020222689 hasConcept C50644808 @default.
• W3020222689 hasConcept C53442348 @default.
• W3020222689 hasConcept C9390403 @default.
• W3020222689 hasConcept C94375191 @default.
• W3020222689 hasConcept C98986596 @default.
• W3020222689 hasConceptScore W3020222689C113775141 @default.
• W3020222689 hasConceptScore W3020222689C119599485 @default.
• W3020222689 hasConceptScore W3020222689C127413603 @default.
• W3020222689 hasConceptScore W3020222689C154945302 @default.
• W3020222689 hasConceptScore W3020222689C165801399 @default.
• W3020222689 hasConceptScore W3020222689C182019814 @default.
• W3020222689 hasConceptScore W3020222689C2778618852 @default.
• W3020222689 hasConceptScore W3020222689C32666082 @default.
• W3020222689 hasConceptScore W3020222689C33923547 @default.
• W3020222689 hasConceptScore W3020222689C41008148 @default.
• W3020222689 hasConceptScore W3020222689C48372109 @default.
• W3020222689 hasConceptScore W3020222689C50644808 @default.
• W3020222689 hasConceptScore W3020222689C53442348 @default.
• W3020222689 hasConceptScore W3020222689C9390403 @default.
• W3020222689 hasConceptScore W3020222689C94375191 @default.
• W3020222689 hasConceptScore W3020222689C98986596 @default.
• W3020222689 hasLocation W30202226891 @default.
• W3020222689 hasLocation W30202226892 @default.
• W3020222689 hasLocation W30202226893 @default.
• W3020222689 hasLocation W30202226894 @default.
• W3020222689 hasLocation W30202226895 @default.
• W3020222689 hasLocation W30202226896 @default.
• W3020222689 hasOpenAccess W3020222689 @default.
• W3020222689 hasPrimaryLocation W30202226891 @default.
• W3020222689 hasRelatedWork W1503438809 @default.
• W3020222689 hasRelatedWork W1675968628 @default.
• W3020222689 hasRelatedWork W1999653134 @default.
• W3020222689 hasRelatedWork W2104200744 @default.
• W3020222689 hasRelatedWork W2136984444 @default.
• W3020222689 hasRelatedWork W2218872326 @default.
• W3020222689 hasRelatedWork W2511463163 @default.
• W3020222689 hasRelatedWork W2911173308 @default.
• W3020222689 hasRelatedWork W3020222689 @default.
• W3020222689 hasRelatedWork W3170645037 @default.
• W3020222689 isParatext "false" @default.
• W3020222689 isRetracted "false" @default.
• W3020222689 magId "3020222689" @default.

• next