FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3104135012> ?p ?o ?g. }

• W3104135012 endingPage "5011" @default.
• W3104135012 startingPage "5004" @default.
• W3104135012 abstract "Current large scale implementations of deep learning and data mining require thousands of processors, massive amounts of off-chip memory, and consume gigajoules of energy. Emerging memory technologies such as nanoscale two-terminal resistive switching memory devices offer a compact, scalable and low power alternative that permits on-chip co-located processing and memory in fine-grain distributed parallel architecture. Here we report first use of resistive switching memory devices for implementing and training a Restricted Boltzmann Machine (RBM), a generative probabilistic graphical model as a key component for unsupervised learning in deep networks. We experimentally demonstrate a 45-synapse RBM realized with 90 resistive switching phase change memory (PCM) elements trained with a bio-inspired variant of the Contrastive Divergence (CD) algorithm, implementing Hebbian and anti-Hebbian weight updates. The resistive PCM devices show a two-fold to ten-fold reduction in error rate in a missing pixel pattern completion task trained over 30 epochs, compared to untrained case. Measured programming energy consumption is 6.1 nJ per epoch with the resistive switching PCM devices, a factor of ~150 times lower than conventional processor-memory systems. We analyze and discuss the dependence of learning performance on cycle-to-cycle variations as well as number of gradual levels in the PCM analog memory devices." @default.
• W3104135012 created "2020-11-23" @default.
• W3104135012 creator A5021876301 @default.
• W3104135012 creator A5043943260 @default.
• W3104135012 creator A5046104015 @default.
• W3104135012 creator A5059013717 @default.
• W3104135012 creator A5060924942 @default.
• W3104135012 creator A5062849115 @default.
• W3104135012 creator A5071086867 @default.
• W3104135012 creator A5088446962 @default.
• W3104135012 creator A5088806568 @default.
• W3104135012 date "2016-12-01" @default.
• W3104135012 modified "2023-10-12" @default.
• W3104135012 title "Training a Probabilistic Graphical Model With Resistive Switching Electronic Synapses" @default.
• W3104135012 cites W1542981317 @default.
• W3104135012 cites W1855112655 @default.
• W3104135012 cites W1967540748 @default.
• W3104135012 cites W1970054482 @default.
• W3104135012 cites W1977773606 @default.
• W3104135012 cites W1981220134 @default.
• W3104135012 cites W1994074248 @default.
• W3104135012 cites W2002700944 @default.
• W3104135012 cites W2030671441 @default.
• W3104135012 cites W2032959979 @default.
• W3104135012 cites W2033597569 @default.
• W3104135012 cites W2064630666 @default.
• W3104135012 cites W2068544601 @default.
• W3104135012 cites W2071947829 @default.
• W3104135012 cites W2081642146 @default.
• W3104135012 cites W2085273892 @default.
• W3104135012 cites W2094043668 @default.
• W3104135012 cites W2094325813 @default.
• W3104135012 cites W2094650137 @default.
• W3104135012 cites W2096192494 @default.
• W3104135012 cites W2100495367 @default.
• W3104135012 cites W2100516830 @default.
• W3104135012 cites W2105103777 @default.
• W3104135012 cites W2107304970 @default.
• W3104135012 cites W2110860124 @default.
• W3104135012 cites W2116064496 @default.
• W3104135012 cites W2117422822 @default.
• W3104135012 cites W2120557145 @default.
• W3104135012 cites W2121208979 @default.
• W3104135012 cites W2130360162 @default.
• W3104135012 cites W2138913040 @default.
• W3104135012 cites W2163922914 @default.
• W3104135012 cites W2194321275 @default.
• W3104135012 cites W2209394967 @default.
• W3104135012 cites W2285660444 @default.
• W3104135012 cites W2288365131 @default.
• W3104135012 cites W2334364695 @default.
• W3104135012 cites W2407840779 @default.
• W3104135012 cites W2919115771 @default.
• W3104135012 cites W3103291989 @default.
• W3104135012 cites W4249616338 @default.
• W3104135012 doi "https://doi.org/10.1109/ted.2016.2616483" @default.
• W3104135012 hasPublicationYear "2016" @default.
• W3104135012 type Work @default.
• W3104135012 sameAs 3104135012 @default.
• W3104135012 citedByCount "35" @default.
• W3104135012 countsByYear W31041350122017 @default.
• W3104135012 countsByYear W31041350122018 @default.
• W3104135012 countsByYear W31041350122019 @default.
• W3104135012 countsByYear W31041350122020 @default.
• W3104135012 countsByYear W31041350122021 @default.
• W3104135012 countsByYear W31041350122022 @default.
• W3104135012 crossrefType "journal-article" @default.
• W3104135012 hasAuthorship W3104135012A5021876301 @default.
• W3104135012 hasAuthorship W3104135012A5043943260 @default.
• W3104135012 hasAuthorship W3104135012A5046104015 @default.
• W3104135012 hasAuthorship W3104135012A5059013717 @default.
• W3104135012 hasAuthorship W3104135012A5060924942 @default.
• W3104135012 hasAuthorship W3104135012A5062849115 @default.
• W3104135012 hasAuthorship W3104135012A5071086867 @default.
• W3104135012 hasAuthorship W3104135012A5088446962 @default.
• W3104135012 hasAuthorship W3104135012A5088806568 @default.
• W3104135012 hasBestOaLocation W31041350122 @default.
• W3104135012 hasConcept C108583219 @default.
• W3104135012 hasConcept C111437709 @default.
• W3104135012 hasConcept C119599485 @default.
• W3104135012 hasConcept C127413603 @default.
• W3104135012 hasConcept C151927369 @default.
• W3104135012 hasConcept C154945302 @default.
• W3104135012 hasConcept C159985019 @default.
• W3104135012 hasConcept C165801399 @default.
• W3104135012 hasConcept C182019814 @default.
• W3104135012 hasConcept C192562407 @default.
• W3104135012 hasConcept C192576344 @default.
• W3104135012 hasConcept C2779227376 @default.
• W3104135012 hasConcept C41008148 @default.
• W3104135012 hasConcept C48044578 @default.
• W3104135012 hasConcept C49937458 @default.
• W3104135012 hasConcept C50644808 @default.
• W3104135012 hasConcept C64142963 @default.
• W3104135012 hasConcept C77088390 @default.
• W3104135012 hasConceptScore W3104135012C108583219 @default.
• W3104135012 hasConceptScore W3104135012C111437709 @default.
• W3104135012 hasConceptScore W3104135012C119599485 @default.

• next