Matches in SemOpenAlex for { <https://semopenalex.org/work/W3202755419> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W3202755419 abstract "Conventional AI accelerators are limited by von-Neumann bottlenecks for edge workloads. Domain-specific accelerators (often neuromorphic) solve this by applying near/in-memory computing, NoC-interconnected massive-multicore setups, and data-flow computation. This requires an effective mapping of neural networks (i.e, an assignment of network layers to cores) to balance resources/memory, computation, and NoC traffic. Here, we introduce a mapping called Snake for the predominant convolutional neural networks (CNNs). It utilizes the feed-forward nature of CNNs by folding layers to spatially adjacent cores. We achieve a total NoC bandwidth improvement of up to 3.8X for MobileNet and ResNet vs. random mappings. Furthermore, NEWROMAP is proposed that continues to optimize Snake mapping through a meta-heuristic; it also simulates the NoC traffic and can work with TensorFlow models. The communication is further optimized with up to 22.52% latency improvement vs. pure snake mapping shown in simulations." @default.
- W3202755419 created "2021-10-11" @default.
- W3202755419 creator A5011744270 @default.
- W3202755419 creator A5023470562 @default.
- W3202755419 creator A5031777933 @default.
- W3202755419 creator A5040881091 @default.
- W3202755419 creator A5043596492 @default.
- W3202755419 date "2021-10-08" @default.
- W3202755419 modified "2023-09-27" @default.
- W3202755419 title "NEWROMAP" @default.
- W3202755419 cites W2161380643 @default.
- W3202755419 cites W2442974303 @default.
- W3202755419 cites W2976973312 @default.
- W3202755419 cites W2985813553 @default.
- W3202755419 cites W3035770626 @default.
- W3202755419 cites W3201876660 @default.
- W3202755419 cites W4244024631 @default.
- W3202755419 doi "https://doi.org/10.1145/3479876.3481591" @default.
- W3202755419 hasPublicationYear "2021" @default.
- W3202755419 type Work @default.
- W3202755419 sameAs 3202755419 @default.
- W3202755419 citedByCount "0" @default.
- W3202755419 crossrefType "proceedings-article" @default.
- W3202755419 hasAuthorship W3202755419A5011744270 @default.
- W3202755419 hasAuthorship W3202755419A5023470562 @default.
- W3202755419 hasAuthorship W3202755419A5031777933 @default.
- W3202755419 hasAuthorship W3202755419A5040881091 @default.
- W3202755419 hasAuthorship W3202755419A5043596492 @default.
- W3202755419 hasConcept C11413529 @default.
- W3202755419 hasConcept C120314980 @default.
- W3202755419 hasConcept C151927369 @default.
- W3202755419 hasConcept C154945302 @default.
- W3202755419 hasConcept C173608175 @default.
- W3202755419 hasConcept C188045654 @default.
- W3202755419 hasConcept C2944601119 @default.
- W3202755419 hasConcept C41008148 @default.
- W3202755419 hasConcept C45374587 @default.
- W3202755419 hasConcept C50644808 @default.
- W3202755419 hasConcept C76155785 @default.
- W3202755419 hasConcept C78766204 @default.
- W3202755419 hasConcept C81363708 @default.
- W3202755419 hasConcept C82876162 @default.
- W3202755419 hasConceptScore W3202755419C11413529 @default.
- W3202755419 hasConceptScore W3202755419C120314980 @default.
- W3202755419 hasConceptScore W3202755419C151927369 @default.
- W3202755419 hasConceptScore W3202755419C154945302 @default.
- W3202755419 hasConceptScore W3202755419C173608175 @default.
- W3202755419 hasConceptScore W3202755419C188045654 @default.
- W3202755419 hasConceptScore W3202755419C2944601119 @default.
- W3202755419 hasConceptScore W3202755419C41008148 @default.
- W3202755419 hasConceptScore W3202755419C45374587 @default.
- W3202755419 hasConceptScore W3202755419C50644808 @default.
- W3202755419 hasConceptScore W3202755419C76155785 @default.
- W3202755419 hasConceptScore W3202755419C78766204 @default.
- W3202755419 hasConceptScore W3202755419C81363708 @default.
- W3202755419 hasConceptScore W3202755419C82876162 @default.
- W3202755419 hasLocation W32027554191 @default.
- W3202755419 hasOpenAccess W3202755419 @default.
- W3202755419 hasPrimaryLocation W32027554191 @default.
- W3202755419 hasRelatedWork W1572523360 @default.
- W3202755419 hasRelatedWork W1966817240 @default.
- W3202755419 hasRelatedWork W1980146226 @default.
- W3202755419 hasRelatedWork W1996757432 @default.
- W3202755419 hasRelatedWork W1997955449 @default.
- W3202755419 hasRelatedWork W2015504232 @default.
- W3202755419 hasRelatedWork W2031026393 @default.
- W3202755419 hasRelatedWork W2086205578 @default.
- W3202755419 hasRelatedWork W2366027386 @default.
- W3202755419 hasRelatedWork W2488897859 @default.
- W3202755419 isParatext "false" @default.
- W3202755419 isRetracted "false" @default.
- W3202755419 magId "3202755419" @default.
- W3202755419 workType "article" @default.