Matches in SemOpenAlex for { <https://semopenalex.org/work/W4246567892> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W4246567892 endingPage "1040" @default.
- W4246567892 startingPage "1040" @default.
- W4246567892 abstract "With the wide deployment of various machine learning algorithms, highly energy-efficient customized machine learning systems have gained popularity. The machine learning compilers are crucial to machine learning systems. The intermediate representation is the key to programming and compilation environments, and it connects the high-level programming language and the lower-level instruction set architectures. The current state-of-the-art intermediate representations are either oriented to high-level algorithms or classical processors based on scalar processing, but they cannot be effectively implemented on tensor-based machine learning systems. To address this problem, we propose a tensor intermediate representation for machine learning systems to improve programming productivity and performance. Concretely, we define a series of tensor types, tensor operations, and tensor memories and optimize the tensor processing based on these definitions. To validate our proposal, we extend the proposed tensor intermediate representation to the low-level scalar intermediate representation of TVM and perform experiments with Tensor Core on a typical machine learning system. Experimental results show that we explore optimizations that are not discovered in the original intermediate representation and achieve 1.62$times$$sim$2.85$times$ performance improvement. Besides, the tensor intermediate representation improves the efficiency of programming by 5.46$times$ on average." @default.
- W4246567892 created "2022-05-12" @default.
- W4246567892 creator A5000432967 @default.
- W4246567892 creator A5063724348 @default.
- W4246567892 creator A5079809798 @default.
- W4246567892 creator A5087827369 @default.
- W4246567892 date "2022-06-01" @default.
- W4246567892 modified "2023-10-16" @default.
- W4246567892 title "A tensor intermediate representation for machine learning systems" @default.
- W4246567892 cites W2000967104 @default.
- W4246567892 cites W2048266589 @default.
- W4246567892 cites W2067523571 @default.
- W4246567892 cites W2152839228 @default.
- W4246567892 cites W2194775991 @default.
- W4246567892 cites W2412412865 @default.
- W4246567892 cites W2442974303 @default.
- W4246567892 cites W2513383847 @default.
- W4246567892 cites W2515287984 @default.
- W4246567892 cites W2531915888 @default.
- W4246567892 cites W2618530766 @default.
- W4246567892 cites W2805566098 @default.
- W4246567892 cites W2945146780 @default.
- W4246567892 cites W2964350391 @default.
- W4246567892 cites W2982083293 @default.
- W4246567892 cites W3101543398 @default.
- W4246567892 cites W3106250896 @default.
- W4246567892 cites W4247198796 @default.
- W4246567892 doi "https://doi.org/10.1360/ssi-2020-0398" @default.
- W4246567892 hasPublicationYear "2022" @default.
- W4246567892 type Work @default.
- W4246567892 citedByCount "0" @default.
- W4246567892 crossrefType "journal-article" @default.
- W4246567892 hasAuthorship W4246567892A5000432967 @default.
- W4246567892 hasAuthorship W4246567892A5063724348 @default.
- W4246567892 hasAuthorship W4246567892A5079809798 @default.
- W4246567892 hasAuthorship W4246567892A5087827369 @default.
- W4246567892 hasConcept C119857082 @default.
- W4246567892 hasConcept C154945302 @default.
- W4246567892 hasConcept C155281189 @default.
- W4246567892 hasConcept C17744445 @default.
- W4246567892 hasConcept C199539241 @default.
- W4246567892 hasConcept C202444582 @default.
- W4246567892 hasConcept C204321447 @default.
- W4246567892 hasConcept C2776359362 @default.
- W4246567892 hasConcept C33923547 @default.
- W4246567892 hasConcept C41008148 @default.
- W4246567892 hasConcept C94625758 @default.
- W4246567892 hasConceptScore W4246567892C119857082 @default.
- W4246567892 hasConceptScore W4246567892C154945302 @default.
- W4246567892 hasConceptScore W4246567892C155281189 @default.
- W4246567892 hasConceptScore W4246567892C17744445 @default.
- W4246567892 hasConceptScore W4246567892C199539241 @default.
- W4246567892 hasConceptScore W4246567892C202444582 @default.
- W4246567892 hasConceptScore W4246567892C204321447 @default.
- W4246567892 hasConceptScore W4246567892C2776359362 @default.
- W4246567892 hasConceptScore W4246567892C33923547 @default.
- W4246567892 hasConceptScore W4246567892C41008148 @default.
- W4246567892 hasConceptScore W4246567892C94625758 @default.
- W4246567892 hasIssue "6" @default.
- W4246567892 hasLocation W42465678921 @default.
- W4246567892 hasOpenAccess W4246567892 @default.
- W4246567892 hasPrimaryLocation W42465678921 @default.
- W4246567892 hasRelatedWork W2611614995 @default.
- W4246567892 hasRelatedWork W2961085424 @default.
- W4246567892 hasRelatedWork W3046775127 @default.
- W4246567892 hasRelatedWork W3107474891 @default.
- W4246567892 hasRelatedWork W4205958290 @default.
- W4246567892 hasRelatedWork W4285260836 @default.
- W4246567892 hasRelatedWork W4286629047 @default.
- W4246567892 hasRelatedWork W4306321456 @default.
- W4246567892 hasRelatedWork W4306674287 @default.
- W4246567892 hasRelatedWork W4224009465 @default.
- W4246567892 hasVolume "52" @default.
- W4246567892 isParatext "false" @default.
- W4246567892 isRetracted "false" @default.
- W4246567892 workType "article" @default.