Matches in SemOpenAlex for { <https://semopenalex.org/work/W3136178412> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W3136178412 abstract "In this paper, we present a novel multi-objective hardware-aware neural architecture search (NAS) framework, namely HSCoNAS, to automate the design of deep neural networks (DNNs) with high accuracy but low latency upon target hardware. To accomplish this goal, we first propose an effective hardware performance modeling method to approximate the runtime latency of DNNs on target hardware, which will be integrated into HSCoNAS to avoid the tedious on-device measurements. Besides, we propose two novel techniques, i.e., dynamic channel scaling to maximize the accuracy under the specified latency and progressive space shrinking to refine the search space towards target hardware as well as alleviate the search overheads. These two techniques jointly work to allow HSCoNAS to perform fine-grained and efficient explorations. Finally, an evolutionary algorithm (EA) is incorporated to conduct the architecture search. Extensive experiments on ImageNet are conducted upon diverse target hardware, i.e., GPU, CPU, and edge device to demonstrate the superiority of HSCoNAS over recent state-of-the-art approaches." @default.
- W3136178412 created "2021-03-29" @default.
- W3136178412 creator A5038598144 @default.
- W3136178412 creator A5039220321 @default.
- W3136178412 creator A5060833244 @default.
- W3136178412 creator A5086240263 @default.
- W3136178412 date "2021-03-11" @default.
- W3136178412 modified "2023-09-25" @default.
- W3136178412 title "HSCoNAS: Hardware-Software Co-Design of Efficient DNNs via Neural Architecture Search" @default.
- W3136178412 cites W2796438033 @default.
- W3136178412 cites W2944779197 @default.
- W3136178412 cites W2947546116 @default.
- W3136178412 cites W2962746461 @default.
- W3136178412 cites W2962856739 @default.
- W3136178412 cites W2963766446 @default.
- W3136178412 cites W2963918968 @default.
- W3136178412 cites W2964081807 @default.
- W3136178412 cites W2964259004 @default.
- W3136178412 cites W2997442807 @default.
- W3136178412 cites W3110617657 @default.
- W3136178412 doi "https://doi.org/10.48550/arxiv.2103.08325" @default.
- W3136178412 hasPublicationYear "2021" @default.
- W3136178412 type Work @default.
- W3136178412 sameAs 3136178412 @default.
- W3136178412 citedByCount "0" @default.
- W3136178412 crossrefType "posted-content" @default.
- W3136178412 hasAuthorship W3136178412A5038598144 @default.
- W3136178412 hasAuthorship W3136178412A5039220321 @default.
- W3136178412 hasAuthorship W3136178412A5060833244 @default.
- W3136178412 hasAuthorship W3136178412A5086240263 @default.
- W3136178412 hasBestOaLocation W31361784121 @default.
- W3136178412 hasConcept C111919701 @default.
- W3136178412 hasConcept C113775141 @default.
- W3136178412 hasConcept C118524514 @default.
- W3136178412 hasConcept C123657996 @default.
- W3136178412 hasConcept C142362112 @default.
- W3136178412 hasConcept C149635348 @default.
- W3136178412 hasConcept C153349607 @default.
- W3136178412 hasConcept C154945302 @default.
- W3136178412 hasConcept C2776221188 @default.
- W3136178412 hasConcept C2777904410 @default.
- W3136178412 hasConcept C2984842247 @default.
- W3136178412 hasConcept C41008148 @default.
- W3136178412 hasConcept C50644808 @default.
- W3136178412 hasConcept C65232700 @default.
- W3136178412 hasConcept C76155785 @default.
- W3136178412 hasConcept C82876162 @default.
- W3136178412 hasConcept C9390403 @default.
- W3136178412 hasConceptScore W3136178412C111919701 @default.
- W3136178412 hasConceptScore W3136178412C113775141 @default.
- W3136178412 hasConceptScore W3136178412C118524514 @default.
- W3136178412 hasConceptScore W3136178412C123657996 @default.
- W3136178412 hasConceptScore W3136178412C142362112 @default.
- W3136178412 hasConceptScore W3136178412C149635348 @default.
- W3136178412 hasConceptScore W3136178412C153349607 @default.
- W3136178412 hasConceptScore W3136178412C154945302 @default.
- W3136178412 hasConceptScore W3136178412C2776221188 @default.
- W3136178412 hasConceptScore W3136178412C2777904410 @default.
- W3136178412 hasConceptScore W3136178412C2984842247 @default.
- W3136178412 hasConceptScore W3136178412C41008148 @default.
- W3136178412 hasConceptScore W3136178412C50644808 @default.
- W3136178412 hasConceptScore W3136178412C65232700 @default.
- W3136178412 hasConceptScore W3136178412C76155785 @default.
- W3136178412 hasConceptScore W3136178412C82876162 @default.
- W3136178412 hasConceptScore W3136178412C9390403 @default.
- W3136178412 hasLocation W31361784121 @default.
- W3136178412 hasLocation W31361784122 @default.
- W3136178412 hasOpenAccess W3136178412 @default.
- W3136178412 hasPrimaryLocation W31361784121 @default.
- W3136178412 hasRelatedWork W2103962833 @default.
- W3136178412 hasRelatedWork W2541819923 @default.
- W3136178412 hasRelatedWork W2780340867 @default.
- W3136178412 hasRelatedWork W3136178412 @default.
- W3136178412 hasRelatedWork W3177213278 @default.
- W3136178412 hasRelatedWork W3186087474 @default.
- W3136178412 hasRelatedWork W3188788977 @default.
- W3136178412 hasRelatedWork W4287273069 @default.
- W3136178412 hasRelatedWork W87354165 @default.
- W3136178412 hasRelatedWork W2114286021 @default.
- W3136178412 isParatext "false" @default.
- W3136178412 isRetracted "false" @default.
- W3136178412 magId "3136178412" @default.
- W3136178412 workType "article" @default.