Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386764810> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W4386764810 abstract "Recent efforts to improve the performance of neural network (NN) accelerators that meet today’s application requirements have given rise to a new trend of logic-based NN inference relying on fixed-function combinational logic (FFCL). This paper presents an innovative optimization methodology for compiling and mapping NNs utilizing FFCL into a logic processor. The presented method maps FFCL blocks to a set of Boolean functions where Boolean operations in each function are mapped to high-performance, low-latency, parallelized processing elements. Graph partitioning and scheduling algorithms are presented to handle FFCL blocks that cannot straightforwardly fit the logic processor. Our experimental evaluations across several datasets and NNs demonstrate the superior performance of our framework in terms of the inference throughput compared to prior art NN accelerators. We achieve 25x higher throughput compared with the XNOR-based accelerator for VGG16 model that can be amplified 5x deploying the graph partitioning and merging algorithms." @default.
- W4386764810 created "2023-09-16" @default.
- W4386764810 creator A5018508596 @default.
- W4386764810 creator A5030070995 @default.
- W4386764810 creator A5044650311 @default.
- W4386764810 creator A5065028689 @default.
- W4386764810 creator A5081336995 @default.
- W4386764810 date "2023-07-09" @default.
- W4386764810 modified "2023-09-27" @default.
- W4386764810 title "Algorithms and Hardware for Efficient Processing of Logic-based Neural Networks" @default.
- W4386764810 cites W2064675550 @default.
- W4386764810 cites W2128921975 @default.
- W4386764810 cites W2908561972 @default.
- W4386764810 cites W2963446712 @default.
- W4386764810 cites W2964137095 @default.
- W4386764810 cites W3008730548 @default.
- W4386764810 cites W3093982621 @default.
- W4386764810 cites W3101493857 @default.
- W4386764810 cites W3102169921 @default.
- W4386764810 cites W3157383824 @default.
- W4386764810 cites W3159008243 @default.
- W4386764810 cites W3205611295 @default.
- W4386764810 doi "https://doi.org/10.1109/dac56929.2023.10247667" @default.
- W4386764810 hasPublicationYear "2023" @default.
- W4386764810 type Work @default.
- W4386764810 citedByCount "0" @default.
- W4386764810 crossrefType "proceedings-article" @default.
- W4386764810 hasAuthorship W4386764810A5018508596 @default.
- W4386764810 hasAuthorship W4386764810A5030070995 @default.
- W4386764810 hasAuthorship W4386764810A5044650311 @default.
- W4386764810 hasAuthorship W4386764810A5065028689 @default.
- W4386764810 hasAuthorship W4386764810A5081336995 @default.
- W4386764810 hasConcept C11413529 @default.
- W4386764810 hasConcept C118524514 @default.
- W4386764810 hasConcept C131017901 @default.
- W4386764810 hasConcept C132525143 @default.
- W4386764810 hasConcept C154945302 @default.
- W4386764810 hasConcept C157922185 @default.
- W4386764810 hasConcept C162324750 @default.
- W4386764810 hasConcept C173608175 @default.
- W4386764810 hasConcept C177264268 @default.
- W4386764810 hasConcept C187455244 @default.
- W4386764810 hasConcept C199360897 @default.
- W4386764810 hasConcept C206729178 @default.
- W4386764810 hasConcept C21547014 @default.
- W4386764810 hasConcept C2776214188 @default.
- W4386764810 hasConcept C41008148 @default.
- W4386764810 hasConcept C50644808 @default.
- W4386764810 hasConcept C80444323 @default.
- W4386764810 hasConcept C81409106 @default.
- W4386764810 hasConceptScore W4386764810C11413529 @default.
- W4386764810 hasConceptScore W4386764810C118524514 @default.
- W4386764810 hasConceptScore W4386764810C131017901 @default.
- W4386764810 hasConceptScore W4386764810C132525143 @default.
- W4386764810 hasConceptScore W4386764810C154945302 @default.
- W4386764810 hasConceptScore W4386764810C157922185 @default.
- W4386764810 hasConceptScore W4386764810C162324750 @default.
- W4386764810 hasConceptScore W4386764810C173608175 @default.
- W4386764810 hasConceptScore W4386764810C177264268 @default.
- W4386764810 hasConceptScore W4386764810C187455244 @default.
- W4386764810 hasConceptScore W4386764810C199360897 @default.
- W4386764810 hasConceptScore W4386764810C206729178 @default.
- W4386764810 hasConceptScore W4386764810C21547014 @default.
- W4386764810 hasConceptScore W4386764810C2776214188 @default.
- W4386764810 hasConceptScore W4386764810C41008148 @default.
- W4386764810 hasConceptScore W4386764810C50644808 @default.
- W4386764810 hasConceptScore W4386764810C80444323 @default.
- W4386764810 hasConceptScore W4386764810C81409106 @default.
- W4386764810 hasFunder F4320306076 @default.
- W4386764810 hasLocation W43867648101 @default.
- W4386764810 hasOpenAccess W4386764810 @default.
- W4386764810 hasPrimaryLocation W43867648101 @default.
- W4386764810 hasRelatedWork W135066275 @default.
- W4386764810 hasRelatedWork W1552215787 @default.
- W4386764810 hasRelatedWork W1843398406 @default.
- W4386764810 hasRelatedWork W1939541994 @default.
- W4386764810 hasRelatedWork W2100323654 @default.
- W4386764810 hasRelatedWork W2116334265 @default.
- W4386764810 hasRelatedWork W2146920875 @default.
- W4386764810 hasRelatedWork W3142612437 @default.
- W4386764810 hasRelatedWork W2001215920 @default.
- W4386764810 hasRelatedWork W2142411491 @default.
- W4386764810 isParatext "false" @default.
- W4386764810 isRetracted "false" @default.
- W4386764810 workType "article" @default.