Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387436176> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4387436176 endingPage "473" @default.
- W4387436176 startingPage "453" @default.
- W4387436176 abstract "As deep learning applications based on convolutional neural networks (CNNs) are increasingly popular in embedded systems, the demand for a customized CNN accelerator, called neural processing unit (NPU), increases since it promises higher performance per watt than a graphics processing unit (GPU). We introduce a hardware–software codesign methodology of a CNN accelerator and present the design of an adder-type NPU, called MIDAP, as an example. The baseline datapath is designed to maximize the utilization of the MAC units by enabling fully pipelined execution of layers without dynamic resource contention. After deciding the baseline datapath, we define a high-level behavior model based on which a high-level compiler and the system simulator are built concurrently. By modifying the simulation models of the hardware units, we could explore the design space of NPU microarchitecture easily. In addition, we could evaluate the effect of hardware extensions to support various types of non-convolutional operations that recent CNN models use widely such as depth-wise convolution and global pooling. After the final datapath is determined, we perform the remaining design steps for NPU implementation: control structure design, low-level compiler design, and RTL synthesis. The end-to-end performance of Inception V3 is expected to be close to 2.0 TOPs/W." @default.
- W4387436176 created "2023-10-09" @default.
- W4387436176 creator A5021206079 @default.
- W4387436176 creator A5029793438 @default.
- W4387436176 date "2023-10-10" @default.
- W4387436176 modified "2023-10-09" @default.
- W4387436176 title "Hardware–Software Codesign of an Adder-Tree Type CNN Accelerator" @default.
- W4387436176 cites W1979486932 @default.
- W4387436176 cites W2034861439 @default.
- W4387436176 cites W3190761184 @default.
- W4387436176 doi "https://doi.org/10.1007/978-3-031-39932-9_18" @default.
- W4387436176 hasPublicationYear "2023" @default.
- W4387436176 type Work @default.
- W4387436176 citedByCount "0" @default.
- W4387436176 crossrefType "book-chapter" @default.
- W4387436176 hasAuthorship W4387436176A5021206079 @default.
- W4387436176 hasAuthorship W4387436176A5029793438 @default.
- W4387436176 hasConcept C118524514 @default.
- W4387436176 hasConcept C149635348 @default.
- W4387436176 hasConcept C154945302 @default.
- W4387436176 hasConcept C164620267 @default.
- W4387436176 hasConcept C169590947 @default.
- W4387436176 hasConcept C173608175 @default.
- W4387436176 hasConcept C199360897 @default.
- W4387436176 hasConcept C2781198647 @default.
- W4387436176 hasConcept C41008148 @default.
- W4387436176 hasConcept C42935608 @default.
- W4387436176 hasConcept C58013763 @default.
- W4387436176 hasConcept C76155785 @default.
- W4387436176 hasConcept C81363708 @default.
- W4387436176 hasConcept C82876162 @default.
- W4387436176 hasConcept C9390403 @default.
- W4387436176 hasConceptScore W4387436176C118524514 @default.
- W4387436176 hasConceptScore W4387436176C149635348 @default.
- W4387436176 hasConceptScore W4387436176C154945302 @default.
- W4387436176 hasConceptScore W4387436176C164620267 @default.
- W4387436176 hasConceptScore W4387436176C169590947 @default.
- W4387436176 hasConceptScore W4387436176C173608175 @default.
- W4387436176 hasConceptScore W4387436176C199360897 @default.
- W4387436176 hasConceptScore W4387436176C2781198647 @default.
- W4387436176 hasConceptScore W4387436176C41008148 @default.
- W4387436176 hasConceptScore W4387436176C42935608 @default.
- W4387436176 hasConceptScore W4387436176C58013763 @default.
- W4387436176 hasConceptScore W4387436176C76155785 @default.
- W4387436176 hasConceptScore W4387436176C81363708 @default.
- W4387436176 hasConceptScore W4387436176C82876162 @default.
- W4387436176 hasConceptScore W4387436176C9390403 @default.
- W4387436176 hasLocation W43874361761 @default.
- W4387436176 hasOpenAccess W4387436176 @default.
- W4387436176 hasPrimaryLocation W43874361761 @default.
- W4387436176 hasRelatedWork W1580556151 @default.
- W4387436176 hasRelatedWork W1839177134 @default.
- W4387436176 hasRelatedWork W1903431847 @default.
- W4387436176 hasRelatedWork W2004001588 @default.
- W4387436176 hasRelatedWork W2084005807 @default.
- W4387436176 hasRelatedWork W2144463068 @default.
- W4387436176 hasRelatedWork W2146736852 @default.
- W4387436176 hasRelatedWork W2166021916 @default.
- W4387436176 hasRelatedWork W2185915791 @default.
- W4387436176 hasRelatedWork W2135482679 @default.
- W4387436176 isParatext "false" @default.
- W4387436176 isRetracted "false" @default.
- W4387436176 workType "book-chapter" @default.