FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2796472956> ?p ?o ?g. }

• W2796472956 endingPage "847" @default.
• W2796472956 startingPage "834" @default.
• W2796472956 abstract "The training of neural networks (NN) is usually time-consuming and resource intensive. The emerging metaloxide resistive random-access memory (RRAM) device has shown potential for the computation of NN. RRAM crossbar structure and multibit characteristics can perform the matrix-vector product in high energy efficiency, which is the most common operation of NN. Two challenges exist for realizing training NN based on RRAM. First, the current architectures based on RRAM only support the inference in training NN and cannot perform the backpropagation (BP) and the weight update of training NN. Second, training NN requires enormous iterations to constantly update the weights for reaching the convergence. However, this weight update leads to large energy consumption because of the nonideal factors of RRAM. In this paper, we propose a training-in-memory based on RRAM (TIME) architecture and the peripheral circuit design to enable training NN on RRAM. TIME supports the BP and the weight update while maximizing the re-usage of peripheral circuits of the inference operation on RRAM. Meanwhile, a set of optimization strategies focusing on the nonideal factors are designed to reduce the cost of tuning RRAM. We explore the performance of both supervised learning (SL) and deep reinforcement learning (DRL) on TIME. A specific mapping method of DRL is also introduced to further improve energy efficiency. Simulation results show that in SL, TIME can achieve 5.3× higher energy efficiency on average compared with DaDianNao, an application-specific integrated circuits (ASIC) in CMOS technology. In DRL, TIME can perform an average 126× higher than GPU in energy efficiency. If the cost of tuning RRAM can be further reduced, TIME has the potential to boost the energy efficiency by two orders of magnitudes compared with ASIC." @default.
• W2796472956 created "2018-04-24" @default.
• W2796472956 creator A5008190519 @default.
• W2796472956 creator A5023755254 @default.
• W2796472956 creator A5036135781 @default.
• W2796472956 creator A5039179408 @default.
• W2796472956 creator A5042532963 @default.
• W2796472956 creator A5068626165 @default.
• W2796472956 creator A5069788729 @default.
• W2796472956 date "2019-05-01" @default.
• W2796472956 modified "2023-10-15" @default.
• W2796472956 title "TIME: A Training-in-Memory Architecture for RRAM-Based Deep Neural Networks" @default.
• W2796472956 cites W1677182931 @default.
• W2796472956 cites W1937359183 @default.
• W2796472956 cites W1973433558 @default.
• W2796472956 cites W1987760418 @default.
• W2796472956 cites W1993163906 @default.
• W2796472956 cites W2013028205 @default.
• W2796472956 cites W2018774711 @default.
• W2796472956 cites W2020740707 @default.
• W2796472956 cites W2024122052 @default.
• W2796472956 cites W2036899386 @default.
• W2796472956 cites W2065076119 @default.
• W2796472956 cites W2080451019 @default.
• W2796472956 cites W2087748124 @default.
• W2796472956 cites W2095076490 @default.
• W2796472956 cites W2108598243 @default.
• W2796472956 cites W2111406701 @default.
• W2796472956 cites W2141546789 @default.
• W2796472956 cites W2145249131 @default.
• W2796472956 cites W2145339207 @default.
• W2796472956 cites W2194775991 @default.
• W2796472956 cites W2248832573 @default.
• W2796472956 cites W2257979135 @default.
• W2796472956 cites W2276486856 @default.
• W2796472956 cites W2346143906 @default.
• W2796472956 cites W2408724663 @default.
• W2796472956 cites W2508602506 @default.
• W2796472956 cites W2517073324 @default.
• W2796472956 cites W2518281301 @default.
• W2796472956 cites W2587907650 @default.
• W2796472956 cites W2588666075 @default.
• W2796472956 cites W2626719825 @default.
• W2796472956 cites W2768104155 @default.
• W2796472956 cites W3099743262 @default.
• W2796472956 cites W4240163901 @default.
• W2796472956 cites W4245795187 @default.
• W2796472956 cites W4251155475 @default.
• W2796472956 doi "https://doi.org/10.1109/tcad.2018.2824304" @default.
• W2796472956 hasPublicationYear "2019" @default.
• W2796472956 type Work @default.
• W2796472956 sameAs 2796472956 @default.
• W2796472956 citedByCount "43" @default.
• W2796472956 countsByYear W27964729562019 @default.
• W2796472956 countsByYear W27964729562020 @default.
• W2796472956 countsByYear W27964729562021 @default.
• W2796472956 countsByYear W27964729562022 @default.
• W2796472956 countsByYear W27964729562023 @default.
• W2796472956 crossrefType "journal-article" @default.
• W2796472956 hasAuthorship W2796472956A5008190519 @default.
• W2796472956 hasAuthorship W2796472956A5023755254 @default.
• W2796472956 hasAuthorship W2796472956A5036135781 @default.
• W2796472956 hasAuthorship W2796472956A5039179408 @default.
• W2796472956 hasAuthorship W2796472956A5042532963 @default.
• W2796472956 hasAuthorship W2796472956A5068626165 @default.
• W2796472956 hasAuthorship W2796472956A5069788729 @default.
• W2796472956 hasConcept C113775141 @default.
• W2796472956 hasConcept C118524514 @default.
• W2796472956 hasConcept C119599485 @default.
• W2796472956 hasConcept C119857082 @default.
• W2796472956 hasConcept C127413603 @default.
• W2796472956 hasConcept C154945302 @default.
• W2796472956 hasConcept C155032097 @default.
• W2796472956 hasConcept C165801399 @default.
• W2796472956 hasConcept C182019814 @default.
• W2796472956 hasConcept C24326235 @default.
• W2796472956 hasConcept C2776214188 @default.
• W2796472956 hasConcept C2780165032 @default.
• W2796472956 hasConcept C41008148 @default.
• W2796472956 hasConcept C50644808 @default.
• W2796472956 hasConceptScore W2796472956C113775141 @default.
• W2796472956 hasConceptScore W2796472956C118524514 @default.
• W2796472956 hasConceptScore W2796472956C119599485 @default.
• W2796472956 hasConceptScore W2796472956C119857082 @default.
• W2796472956 hasConceptScore W2796472956C127413603 @default.
• W2796472956 hasConceptScore W2796472956C154945302 @default.
• W2796472956 hasConceptScore W2796472956C155032097 @default.
• W2796472956 hasConceptScore W2796472956C165801399 @default.
• W2796472956 hasConceptScore W2796472956C182019814 @default.
• W2796472956 hasConceptScore W2796472956C24326235 @default.
• W2796472956 hasConceptScore W2796472956C2776214188 @default.
• W2796472956 hasConceptScore W2796472956C2780165032 @default.
• W2796472956 hasConceptScore W2796472956C41008148 @default.
• W2796472956 hasConceptScore W2796472956C50644808 @default.
• W2796472956 hasFunder F4320321001 @default.
• W2796472956 hasFunder F4320335777 @default.
• W2796472956 hasIssue "5" @default.
• W2796472956 hasLocation W27964729561 @default.

• next