Matches in SemOpenAlex for { <https://semopenalex.org/work/W2615398246> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W2615398246 endingPage "19" @default.
- W2615398246 startingPage "1" @default.
- W2615398246 abstract "Neural network (NN) systems are widely used in many important applications ranging from computer vision to speech recognition. To date, most NN systems are processed by general processing units like CPUs or GPUs. However, as the sizes of dataset and network rapidly increase, the original software implementations suffer from long training time. To overcome this problem, specialized hardware accelerators are needed to design high-speed NN systems. This article presents an efficient hardware architecture of restricted Boltzmann machine (RBM) that is an important category of NN systems. Various optimization approaches at the hardware level are performed to improve the training speed. As-soon-as-possible and overlapped-scheduling approaches are used to reduce the latency. It is shown that, compared with the flat design, the proposed RBM architecture can achieve 50% reduction in training time. In addition, an on-the-fly computation scheme is also used to reduce the storage requirement of binary and stochastic states by several hundreds of times. Then, based on the proposed approach, a 784-2252 RBM design example is developed for MNIST handwritten digit recognition dataset. Analysis shows that the VLSI design of RBM achieves significant improvement in training speed and energy efficiency as compared to CPU/GPU-based solution." @default.
- W2615398246 created "2017-05-26" @default.
- W2615398246 creator A5007884053 @default.
- W2615398246 creator A5083854815 @default.
- W2615398246 date "2017-05-12" @default.
- W2615398246 modified "2023-09-23" @default.
- W2615398246 title "VLSI Architectures for the Restricted Boltzmann Machine" @default.
- W2615398246 cites W1524093576 @default.
- W2615398246 cites W1890351280 @default.
- W2615398246 cites W1968422655 @default.
- W2615398246 cites W1973695593 @default.
- W2615398246 cites W1992286510 @default.
- W2615398246 cites W2009832130 @default.
- W2615398246 cites W2099736035 @default.
- W2615398246 cites W2108379813 @default.
- W2615398246 cites W2115448218 @default.
- W2615398246 cites W2121423846 @default.
- W2615398246 cites W2126838646 @default.
- W2615398246 cites W2136922672 @default.
- W2615398246 cites W2150609304 @default.
- W2615398246 cites W2152839228 @default.
- W2615398246 cites W2171318387 @default.
- W2615398246 doi "https://doi.org/10.1145/3007193" @default.
- W2615398246 hasPublicationYear "2017" @default.
- W2615398246 type Work @default.
- W2615398246 sameAs 2615398246 @default.
- W2615398246 citedByCount "5" @default.
- W2615398246 countsByYear W26153982462019 @default.
- W2615398246 countsByYear W26153982462020 @default.
- W2615398246 countsByYear W26153982462021 @default.
- W2615398246 crossrefType "journal-article" @default.
- W2615398246 hasAuthorship W2615398246A5007884053 @default.
- W2615398246 hasAuthorship W2615398246A5083854815 @default.
- W2615398246 hasConcept C113775141 @default.
- W2615398246 hasConcept C11413529 @default.
- W2615398246 hasConcept C118524514 @default.
- W2615398246 hasConcept C14580979 @default.
- W2615398246 hasConcept C149635348 @default.
- W2615398246 hasConcept C154945302 @default.
- W2615398246 hasConcept C162324750 @default.
- W2615398246 hasConcept C173608175 @default.
- W2615398246 hasConcept C190502265 @default.
- W2615398246 hasConcept C192576344 @default.
- W2615398246 hasConcept C199354608 @default.
- W2615398246 hasConcept C206729178 @default.
- W2615398246 hasConcept C21547014 @default.
- W2615398246 hasConcept C41008148 @default.
- W2615398246 hasConcept C42935608 @default.
- W2615398246 hasConcept C45374587 @default.
- W2615398246 hasConcept C50644808 @default.
- W2615398246 hasConcept C9390403 @default.
- W2615398246 hasConceptScore W2615398246C113775141 @default.
- W2615398246 hasConceptScore W2615398246C11413529 @default.
- W2615398246 hasConceptScore W2615398246C118524514 @default.
- W2615398246 hasConceptScore W2615398246C14580979 @default.
- W2615398246 hasConceptScore W2615398246C149635348 @default.
- W2615398246 hasConceptScore W2615398246C154945302 @default.
- W2615398246 hasConceptScore W2615398246C162324750 @default.
- W2615398246 hasConceptScore W2615398246C173608175 @default.
- W2615398246 hasConceptScore W2615398246C190502265 @default.
- W2615398246 hasConceptScore W2615398246C192576344 @default.
- W2615398246 hasConceptScore W2615398246C199354608 @default.
- W2615398246 hasConceptScore W2615398246C206729178 @default.
- W2615398246 hasConceptScore W2615398246C21547014 @default.
- W2615398246 hasConceptScore W2615398246C41008148 @default.
- W2615398246 hasConceptScore W2615398246C42935608 @default.
- W2615398246 hasConceptScore W2615398246C45374587 @default.
- W2615398246 hasConceptScore W2615398246C50644808 @default.
- W2615398246 hasConceptScore W2615398246C9390403 @default.
- W2615398246 hasIssue "3" @default.
- W2615398246 hasLocation W26153982461 @default.
- W2615398246 hasOpenAccess W2615398246 @default.
- W2615398246 hasPrimaryLocation W26153982461 @default.
- W2615398246 hasRelatedWork W1860547750 @default.
- W2615398246 hasRelatedWork W2062196006 @default.
- W2615398246 hasRelatedWork W2369375926 @default.
- W2615398246 hasRelatedWork W2461687944 @default.
- W2615398246 hasRelatedWork W2615398246 @default.
- W2615398246 hasRelatedWork W3042265279 @default.
- W2615398246 hasRelatedWork W3124648670 @default.
- W2615398246 hasRelatedWork W3217159865 @default.
- W2615398246 hasRelatedWork W4214872881 @default.
- W2615398246 hasRelatedWork W2046957099 @default.
- W2615398246 hasVolume "13" @default.
- W2615398246 isParatext "false" @default.
- W2615398246 isRetracted "false" @default.
- W2615398246 magId "2615398246" @default.
- W2615398246 workType "article" @default.