FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3035226078> ?p ?o ?g. }

• W3035226078 endingPage "2305" @default.
• W3035226078 startingPage "2293" @default.
• W3035226078 abstract "The spread of deep learning on embedded devices has prompted the development of numerous methods to optimize the deployment of deep neural networks (DNNs). Works have mainly focused on: 1) efficient DNN architectures; 2) network optimization techniques, such as pruning and quantization; 3) optimized algorithms to speed up the execution of the most computational intensive layers; and 4) dedicated hardware to accelerate the data flow and computation. However, there is a lack of research on cross-level optimization as the space of approaches becomes too large to test and obtain a globally optimized solution. Thus, leading to suboptimal deployment in terms of latency, accuracy, and memory. In this work, we first detail and analyze the methods to improve the deployment of DNNs across the different levels of software optimization. Building on this knowledge, we present an automated exploration framework to ease the deployment of DNNs. The framework relies on a reinforcement learning search that, combined with a deep learning inference framework, automatically explores the design space and learns an optimized solution that speeds up the performance and reduces the memory on embedded CPU platforms. Thus, we present a set of results for state-of-the-art DNNs on a range of Arm Cortex-A CPU platforms achieving up to <inline-formula xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink> <tex-math notation=LaTeX>$4times $ </tex-math></inline-formula> improvement in performance and over <inline-formula xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink> <tex-math notation=LaTeX>$2times $ </tex-math></inline-formula> reduction in memory with negligible loss in accuracy with respect to the BLAS floating-point implementation." @default.
• W3035226078 created "2020-06-19" @default.
• W3035226078 creator A5005704374 @default.
• W3035226078 creator A5028033266 @default.
• W3035226078 creator A5036407549 @default.
• W3035226078 creator A5039374801 @default.
• W3035226078 creator A5043680067 @default.
• W3035226078 creator A5082631490 @default.
• W3035226078 date "2021-11-01" @default.
• W3035226078 modified "2023-09-24" @default.
• W3035226078 title "Automated Design Space Exploration for Optimized Deployment of DNN on Arm Cortex-A CPUs" @default.
• W3035226078 cites W1969067576 @default.
• W3035226078 cites W1969483458 @default.
• W3035226078 cites W2094756095 @default.
• W3035226078 cites W2141559645 @default.
• W3035226078 cites W2143612262 @default.
• W3035226078 cites W2145339207 @default.
• W3035226078 cites W2169528473 @default.
• W3035226078 cites W2172654076 @default.
• W3035226078 cites W2217896605 @default.
• W3035226078 cites W2252007067 @default.
• W3035226078 cites W2276892413 @default.
• W3035226078 cites W2300242332 @default.
• W3035226078 cites W2431931973 @default.
• W3035226078 cites W2583383421 @default.
• W3035226078 cites W2596075132 @default.
• W3035226078 cites W2621749091 @default.
• W3035226078 cites W2781752755 @default.
• W3035226078 cites W2809254203 @default.
• W3035226078 cites W2886851211 @default.
• W3035226078 cites W2901177033 @default.
• W3035226078 cites W2931743911 @default.
• W3035226078 cites W2962861284 @default.
• W3035226078 cites W2963122961 @default.
• W3035226078 cites W2963125010 @default.
• W3035226078 cites W2963163009 @default.
• W3035226078 cites W2963918968 @default.
• W3035226078 cites W2967733054 @default.
• W3035226078 cites W2981751377 @default.
• W3035226078 cites W2982041622 @default.
• W3035226078 cites W2982479999 @default.
• W3035226078 cites W3011121930 @default.
• W3035226078 cites W3099541711 @default.
• W3035226078 cites W4288083474 @default.
• W3035226078 doi "https://doi.org/10.1109/tcad.2020.3046568" @default.
• W3035226078 hasPublicationYear "2021" @default.
• W3035226078 type Work @default.
• W3035226078 sameAs 3035226078 @default.
• W3035226078 citedByCount "6" @default.
• W3035226078 countsByYear W30352260782021 @default.
• W3035226078 countsByYear W30352260782022 @default.
• W3035226078 countsByYear W30352260782023 @default.
• W3035226078 crossrefType "journal-article" @default.
• W3035226078 hasAuthorship W3035226078A5005704374 @default.
• W3035226078 hasAuthorship W3035226078A5028033266 @default.
• W3035226078 hasAuthorship W3035226078A5036407549 @default.
• W3035226078 hasAuthorship W3035226078A5039374801 @default.
• W3035226078 hasAuthorship W3035226078A5043680067 @default.
• W3035226078 hasAuthorship W3035226078A5082631490 @default.
• W3035226078 hasBestOaLocation W30352260782 @default.
• W3035226078 hasConcept C105339364 @default.
• W3035226078 hasConcept C108010975 @default.
• W3035226078 hasConcept C108583219 @default.
• W3035226078 hasConcept C111919701 @default.
• W3035226078 hasConcept C113775141 @default.
• W3035226078 hasConcept C118524514 @default.
• W3035226078 hasConcept C119857082 @default.
• W3035226078 hasConcept C149635348 @default.
• W3035226078 hasConcept C154945302 @default.
• W3035226078 hasConcept C173608175 @default.
• W3035226078 hasConcept C2776214188 @default.
• W3035226078 hasConcept C2776221188 @default.
• W3035226078 hasConcept C41008148 @default.
• W3035226078 hasConcept C50644808 @default.
• W3035226078 hasConcept C6557445 @default.
• W3035226078 hasConcept C76155785 @default.
• W3035226078 hasConcept C82876162 @default.
• W3035226078 hasConcept C86803240 @default.
• W3035226078 hasConcept C97541855 @default.
• W3035226078 hasConceptScore W3035226078C105339364 @default.
• W3035226078 hasConceptScore W3035226078C108010975 @default.
• W3035226078 hasConceptScore W3035226078C108583219 @default.
• W3035226078 hasConceptScore W3035226078C111919701 @default.
• W3035226078 hasConceptScore W3035226078C113775141 @default.
• W3035226078 hasConceptScore W3035226078C118524514 @default.
• W3035226078 hasConceptScore W3035226078C119857082 @default.
• W3035226078 hasConceptScore W3035226078C149635348 @default.
• W3035226078 hasConceptScore W3035226078C154945302 @default.
• W3035226078 hasConceptScore W3035226078C173608175 @default.
• W3035226078 hasConceptScore W3035226078C2776214188 @default.
• W3035226078 hasConceptScore W3035226078C2776221188 @default.
• W3035226078 hasConceptScore W3035226078C41008148 @default.
• W3035226078 hasConceptScore W3035226078C50644808 @default.
• W3035226078 hasConceptScore W3035226078C6557445 @default.
• W3035226078 hasConceptScore W3035226078C76155785 @default.
• W3035226078 hasConceptScore W3035226078C82876162 @default.
• W3035226078 hasConceptScore W3035226078C86803240 @default.
• W3035226078 hasConceptScore W3035226078C97541855 @default.

• next