Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897366704> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2897366704 abstract "Specialized ultra-low power deep learning architectures with on-chip training capability can be useful in variety of applications that require adaptability. This paper presents such a processor design, Socrates-D 2.0, a multicore architecture for deep neural network based training and inference. The architecture consists of a set of processing cores, each with internal memories to store synaptic weights. Additionally, we present a method to map traditional deep learning networks to our multicore architecture and show that there is minimal impact in training accuracy. The system level area and power benefits of the specialized architecture are compared with the earlier generation of Socrates-D. Our experimental evaluations show that the proposed architecture can provide 1.25× area and 1.19× energy efficiency than the previous version of Socrates-D." @default.
- W2897366704 created "2018-10-26" @default.
- W2897366704 creator A5021250695 @default.
- W2897366704 creator A5025518911 @default.
- W2897366704 creator A5048372156 @default.
- W2897366704 date "2018-07-01" @default.
- W2897366704 modified "2023-10-16" @default.
- W2897366704 title "Socrates-D 2.0: A Low Power High Throughput Architecture for Deep Network Training" @default.
- W2897366704 cites W2023745373 @default.
- W2897366704 cites W2047223359 @default.
- W2897366704 cites W2048266589 @default.
- W2897366704 cites W2120480077 @default.
- W2897366704 cites W2122410182 @default.
- W2897366704 cites W2138913040 @default.
- W2897366704 cites W2170382128 @default.
- W2897366704 cites W2285660444 @default.
- W2897366704 cites W2289252105 @default.
- W2897366704 cites W2508602506 @default.
- W2897366704 cites W2511743527 @default.
- W2897366704 cites W2518281301 @default.
- W2897366704 cites W2518511512 @default.
- W2897366704 cites W2560912757 @default.
- W2897366704 cites W2772992092 @default.
- W2897366704 doi "https://doi.org/10.1109/ijcnn.2018.8489378" @default.
- W2897366704 hasPublicationYear "2018" @default.
- W2897366704 type Work @default.
- W2897366704 sameAs 2897366704 @default.
- W2897366704 citedByCount "0" @default.
- W2897366704 crossrefType "proceedings-article" @default.
- W2897366704 hasAuthorship W2897366704A5021250695 @default.
- W2897366704 hasAuthorship W2897366704A5025518911 @default.
- W2897366704 hasAuthorship W2897366704A5048372156 @default.
- W2897366704 hasConcept C108583219 @default.
- W2897366704 hasConcept C111472728 @default.
- W2897366704 hasConcept C118524514 @default.
- W2897366704 hasConcept C123657996 @default.
- W2897366704 hasConcept C138885662 @default.
- W2897366704 hasConcept C142362112 @default.
- W2897366704 hasConcept C149635348 @default.
- W2897366704 hasConcept C153349607 @default.
- W2897366704 hasConcept C154945302 @default.
- W2897366704 hasConcept C157764524 @default.
- W2897366704 hasConcept C173608175 @default.
- W2897366704 hasConcept C177606310 @default.
- W2897366704 hasConcept C18903297 @default.
- W2897366704 hasConcept C193415008 @default.
- W2897366704 hasConcept C2776214188 @default.
- W2897366704 hasConcept C2779259174 @default.
- W2897366704 hasConcept C31258907 @default.
- W2897366704 hasConcept C41008148 @default.
- W2897366704 hasConcept C50644808 @default.
- W2897366704 hasConcept C555944384 @default.
- W2897366704 hasConcept C76155785 @default.
- W2897366704 hasConcept C78766204 @default.
- W2897366704 hasConcept C86803240 @default.
- W2897366704 hasConceptScore W2897366704C108583219 @default.
- W2897366704 hasConceptScore W2897366704C111472728 @default.
- W2897366704 hasConceptScore W2897366704C118524514 @default.
- W2897366704 hasConceptScore W2897366704C123657996 @default.
- W2897366704 hasConceptScore W2897366704C138885662 @default.
- W2897366704 hasConceptScore W2897366704C142362112 @default.
- W2897366704 hasConceptScore W2897366704C149635348 @default.
- W2897366704 hasConceptScore W2897366704C153349607 @default.
- W2897366704 hasConceptScore W2897366704C154945302 @default.
- W2897366704 hasConceptScore W2897366704C157764524 @default.
- W2897366704 hasConceptScore W2897366704C173608175 @default.
- W2897366704 hasConceptScore W2897366704C177606310 @default.
- W2897366704 hasConceptScore W2897366704C18903297 @default.
- W2897366704 hasConceptScore W2897366704C193415008 @default.
- W2897366704 hasConceptScore W2897366704C2776214188 @default.
- W2897366704 hasConceptScore W2897366704C2779259174 @default.
- W2897366704 hasConceptScore W2897366704C31258907 @default.
- W2897366704 hasConceptScore W2897366704C41008148 @default.
- W2897366704 hasConceptScore W2897366704C50644808 @default.
- W2897366704 hasConceptScore W2897366704C555944384 @default.
- W2897366704 hasConceptScore W2897366704C76155785 @default.
- W2897366704 hasConceptScore W2897366704C78766204 @default.
- W2897366704 hasConceptScore W2897366704C86803240 @default.
- W2897366704 hasLocation W28973667041 @default.
- W2897366704 hasOpenAccess W2897366704 @default.
- W2897366704 hasPrimaryLocation W28973667041 @default.
- W2897366704 hasRelatedWork W2312941002 @default.
- W2897366704 hasRelatedWork W2352173818 @default.
- W2897366704 hasRelatedWork W2364531466 @default.
- W2897366704 hasRelatedWork W2382623646 @default.
- W2897366704 hasRelatedWork W2544929033 @default.
- W2897366704 hasRelatedWork W2783121225 @default.
- W2897366704 hasRelatedWork W2897366704 @default.
- W2897366704 hasRelatedWork W4200276911 @default.
- W2897366704 hasRelatedWork W4206841102 @default.
- W2897366704 hasRelatedWork W642962640 @default.
- W2897366704 isParatext "false" @default.
- W2897366704 isRetracted "false" @default.
- W2897366704 magId "2897366704" @default.
- W2897366704 workType "article" @default.