Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385804961> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W4385804961 abstract "Low-Power Edge-AI capabilities are essential for on-device extended reality (XR) applications to support the vision of Metaverse. A critical requirement for emerging AI applications is personalization and adaptability without requiring retraining. Few-shot learning using embedding-based computations present an attractive method for the same. However, quantization-based optimizations to map such computations are yet to be explored. In this work, we present a fully binarized distance computing (BinDC) framework to perform distance computations for few-shot learning using only accumulation and logic operations (XOR/XNOR). The proposed method leads to marginal loss in accuracy of ≈ 4% (for 4-bits). This leads to savings in memory (≈ 8 ), energy (≈ 2.5-3×), power (≈ 2×) and latency (≈ 1.1-1.5×) compared to a floating-point cosine distance computation when using CPU-based computations performed on an embedded platform. We further demonstrate realization utilizing RRAM (resistive random access memory) based IMC (in-memory computing) to further improve EDP (energy delay product) (≈ 1000×) in comparison to the embedded CPU-based realization." @default.
- W4385804961 created "2023-08-15" @default.
- W4385804961 creator A5008011360 @default.
- W4385804961 creator A5053194542 @default.
- W4385804961 creator A5066176656 @default.
- W4385804961 creator A5072122272 @default.
- W4385804961 creator A5074593268 @default.
- W4385804961 creator A5088965878 @default.
- W4385804961 date "2023-06-01" @default.
- W4385804961 modified "2023-10-18" @default.
- W4385804961 title "Fully-Binarized Distance Computation based On-device Few-Shot Learning for XR applications" @default.
- W4385804961 cites W2108598243 @default.
- W4385804961 cites W2895774650 @default.
- W4385804961 cites W2988640543 @default.
- W4385804961 cites W3010691265 @default.
- W4385804961 cites W3032819016 @default.
- W4385804961 cites W3034312118 @default.
- W4385804961 cites W3171572324 @default.
- W4385804961 cites W3200699344 @default.
- W4385804961 cites W3200825586 @default.
- W4385804961 cites W4292793932 @default.
- W4385804961 cites W4296209087 @default.
- W4385804961 cites W4308089783 @default.
- W4385804961 doi "https://doi.org/10.1109/cvprw59228.2023.00473" @default.
- W4385804961 hasPublicationYear "2023" @default.
- W4385804961 type Work @default.
- W4385804961 citedByCount "0" @default.
- W4385804961 crossrefType "proceedings-article" @default.
- W4385804961 hasAuthorship W4385804961A5008011360 @default.
- W4385804961 hasAuthorship W4385804961A5053194542 @default.
- W4385804961 hasAuthorship W4385804961A5066176656 @default.
- W4385804961 hasAuthorship W4385804961A5072122272 @default.
- W4385804961 hasAuthorship W4385804961A5074593268 @default.
- W4385804961 hasAuthorship W4385804961A5088965878 @default.
- W4385804961 hasConcept C105795698 @default.
- W4385804961 hasConcept C113775141 @default.
- W4385804961 hasConcept C11413529 @default.
- W4385804961 hasConcept C124296912 @default.
- W4385804961 hasConcept C131017901 @default.
- W4385804961 hasConcept C154945302 @default.
- W4385804961 hasConcept C162307627 @default.
- W4385804961 hasConcept C2778456923 @default.
- W4385804961 hasConcept C2781089630 @default.
- W4385804961 hasConcept C28855332 @default.
- W4385804961 hasConcept C33923547 @default.
- W4385804961 hasConcept C41008148 @default.
- W4385804961 hasConcept C41608201 @default.
- W4385804961 hasConcept C45374587 @default.
- W4385804961 hasConcept C57684291 @default.
- W4385804961 hasConcept C9390403 @default.
- W4385804961 hasConceptScore W4385804961C105795698 @default.
- W4385804961 hasConceptScore W4385804961C113775141 @default.
- W4385804961 hasConceptScore W4385804961C11413529 @default.
- W4385804961 hasConceptScore W4385804961C124296912 @default.
- W4385804961 hasConceptScore W4385804961C131017901 @default.
- W4385804961 hasConceptScore W4385804961C154945302 @default.
- W4385804961 hasConceptScore W4385804961C162307627 @default.
- W4385804961 hasConceptScore W4385804961C2778456923 @default.
- W4385804961 hasConceptScore W4385804961C2781089630 @default.
- W4385804961 hasConceptScore W4385804961C28855332 @default.
- W4385804961 hasConceptScore W4385804961C33923547 @default.
- W4385804961 hasConceptScore W4385804961C41008148 @default.
- W4385804961 hasConceptScore W4385804961C41608201 @default.
- W4385804961 hasConceptScore W4385804961C45374587 @default.
- W4385804961 hasConceptScore W4385804961C57684291 @default.
- W4385804961 hasConceptScore W4385804961C9390403 @default.
- W4385804961 hasLocation W43858049611 @default.
- W4385804961 hasOpenAccess W4385804961 @default.
- W4385804961 hasPrimaryLocation W43858049611 @default.
- W4385804961 hasRelatedWork W1519132907 @default.
- W4385804961 hasRelatedWork W2055709700 @default.
- W4385804961 hasRelatedWork W2351210157 @default.
- W4385804961 hasRelatedWork W2354485250 @default.
- W4385804961 hasRelatedWork W2354943544 @default.
- W4385804961 hasRelatedWork W2973622361 @default.
- W4385804961 hasRelatedWork W3008102594 @default.
- W4385804961 hasRelatedWork W3135013065 @default.
- W4385804961 hasRelatedWork W3216454443 @default.
- W4385804961 hasRelatedWork W4312069028 @default.
- W4385804961 isParatext "false" @default.
- W4385804961 isRetracted "false" @default.
- W4385804961 workType "article" @default.