Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384835336> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4384835336 abstract "Quantized deep learning models are suitable for the embedded devices with limited computation resource. For computation-intensive neural network operators such as convolution, heterogeneous platforms with a set of processing units of different types become common in the embedded devices. These embedded devices usually operate on fixed-point calculations; moreover, they rely on customized kernel functions to deploy deep learning models. In this paper, a flow of deploying pre-quantized deep learning models on heterogeneous platforms using TVM is presented. We propose an optimization to convert quantization parameters. To leverage customized kernel functions, we propose the operator flow recognition. To demonstrate our flow, we utilize embARC Machine Learning Inference (embARC MLI), an open-source software library targeted for low-power applications. A set of pre-quantized deep learning models are deployed on a heterogeneous platform comprising x86 and embARC MLI. Experimental results show that for each model, the accuracy obtained from the heterogeneous platform is much the same as the one obtained from an x86 platform." @default.
- W4384835336 created "2023-07-21" @default.
- W4384835336 creator A5015376105 @default.
- W4384835336 creator A5039993264 @default.
- W4384835336 creator A5071820714 @default.
- W4384835336 creator A5091000621 @default.
- W4384835336 date "2023-04-21" @default.
- W4384835336 modified "2023-09-23" @default.
- W4384835336 title "Deploying Pre-Quantized Deep Learning Models on Heterogeneous Platforms with Operator Flow Recognition and Quantization Parameter Optimization" @default.
- W4384835336 cites W2194775991 @default.
- W4384835336 cites W2476548250 @default.
- W4384835336 cites W2805566098 @default.
- W4384835336 cites W2963122961 @default.
- W4384835336 cites W2970514904 @default.
- W4384835336 cites W4214951654 @default.
- W4384835336 doi "https://doi.org/10.1109/icasi57738.2023.10179562" @default.
- W4384835336 hasPublicationYear "2023" @default.
- W4384835336 type Work @default.
- W4384835336 citedByCount "0" @default.
- W4384835336 crossrefType "proceedings-article" @default.
- W4384835336 hasAuthorship W4384835336A5015376105 @default.
- W4384835336 hasAuthorship W4384835336A5039993264 @default.
- W4384835336 hasAuthorship W4384835336A5071820714 @default.
- W4384835336 hasAuthorship W4384835336A5091000621 @default.
- W4384835336 hasConcept C108583219 @default.
- W4384835336 hasConcept C113775141 @default.
- W4384835336 hasConcept C11413529 @default.
- W4384835336 hasConcept C114614502 @default.
- W4384835336 hasConcept C119857082 @default.
- W4384835336 hasConcept C153083717 @default.
- W4384835336 hasConcept C154945302 @default.
- W4384835336 hasConcept C170723468 @default.
- W4384835336 hasConcept C199360897 @default.
- W4384835336 hasConcept C2776214188 @default.
- W4384835336 hasConcept C2777904410 @default.
- W4384835336 hasConcept C28855332 @default.
- W4384835336 hasConcept C33923547 @default.
- W4384835336 hasConcept C41008148 @default.
- W4384835336 hasConcept C45374587 @default.
- W4384835336 hasConcept C50644808 @default.
- W4384835336 hasConcept C74193536 @default.
- W4384835336 hasConceptScore W4384835336C108583219 @default.
- W4384835336 hasConceptScore W4384835336C113775141 @default.
- W4384835336 hasConceptScore W4384835336C11413529 @default.
- W4384835336 hasConceptScore W4384835336C114614502 @default.
- W4384835336 hasConceptScore W4384835336C119857082 @default.
- W4384835336 hasConceptScore W4384835336C153083717 @default.
- W4384835336 hasConceptScore W4384835336C154945302 @default.
- W4384835336 hasConceptScore W4384835336C170723468 @default.
- W4384835336 hasConceptScore W4384835336C199360897 @default.
- W4384835336 hasConceptScore W4384835336C2776214188 @default.
- W4384835336 hasConceptScore W4384835336C2777904410 @default.
- W4384835336 hasConceptScore W4384835336C28855332 @default.
- W4384835336 hasConceptScore W4384835336C33923547 @default.
- W4384835336 hasConceptScore W4384835336C41008148 @default.
- W4384835336 hasConceptScore W4384835336C45374587 @default.
- W4384835336 hasConceptScore W4384835336C50644808 @default.
- W4384835336 hasConceptScore W4384835336C74193536 @default.
- W4384835336 hasLocation W43848353361 @default.
- W4384835336 hasOpenAccess W4384835336 @default.
- W4384835336 hasPrimaryLocation W43848353361 @default.
- W4384835336 hasRelatedWork W2963058055 @default.
- W4384835336 hasRelatedWork W3131807303 @default.
- W4384835336 hasRelatedWork W3133174490 @default.
- W4384835336 hasRelatedWork W4223943233 @default.
- W4384835336 hasRelatedWork W4309045103 @default.
- W4384835336 hasRelatedWork W4312200629 @default.
- W4384835336 hasRelatedWork W4323922040 @default.
- W4384835336 hasRelatedWork W4360585206 @default.
- W4384835336 hasRelatedWork W4364306694 @default.
- W4384835336 hasRelatedWork W4380086463 @default.
- W4384835336 isParatext "false" @default.
- W4384835336 isRetracted "false" @default.
- W4384835336 workType "article" @default.