Matches in SemOpenAlex for { <https://semopenalex.org/work/W3092611426> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W3092611426 abstract "Reducing the model size and computation costs for dedicated AI accelerator designs, neural network quantization methods have attracted momentous attention recently. Unfortunately, merely minimizing quantization loss using constant discretization causes accuracy deterioration. In this paper, we propose an iterative accuracy-driven learning framework of competitive-collaborative quantization (CCQ) to gradually adapt the bit-precision of each individual layer. Orthogonal to prior quantization policies working with full precision for the first and last layers of the network, CCQ offers layer-wise competition for any target quantization policy with holistic layer fine-tuning to recover accuracy, where the state-of-the-art networks can be entirely quantized without any significant accuracy degradation." @default.
- W3092611426 created "2020-10-15" @default.
- W3092611426 creator A5015236818 @default.
- W3092611426 creator A5038935739 @default.
- W3092611426 creator A5076019669 @default.
- W3092611426 creator A5076083402 @default.
- W3092611426 date "2020-07-01" @default.
- W3092611426 modified "2023-09-26" @default.
- W3092611426 title "Learning to Quantize Deep Neural Networks: A Competitive-Collaborative Approach" @default.
- W3092611426 cites W2077723394 @default.
- W3092611426 cites W2108598243 @default.
- W3092611426 cites W2194775991 @default.
- W3092611426 cites W2267635276 @default.
- W3092611426 cites W2300242332 @default.
- W3092611426 cites W2319920447 @default.
- W3092611426 cites W2469490737 @default.
- W3092611426 cites W2518108298 @default.
- W3092611426 cites W2653504903 @default.
- W3092611426 cites W2786771851 @default.
- W3092611426 cites W2798332427 @default.
- W3092611426 cites W2809371234 @default.
- W3092611426 cites W2884150179 @default.
- W3092611426 cites W2886847195 @default.
- W3092611426 cites W2896718327 @default.
- W3092611426 cites W2913190747 @default.
- W3092611426 cites W2946787800 @default.
- W3092611426 cites W2947515409 @default.
- W3092611426 cites W2962856739 @default.
- W3092611426 cites W2963480671 @default.
- W3092611426 cites W2963547822 @default.
- W3092611426 cites W2970020383 @default.
- W3092611426 cites W2982041622 @default.
- W3092611426 cites W2982479999 @default.
- W3092611426 cites W2982654255 @default.
- W3092611426 cites W2994840239 @default.
- W3092611426 cites W3118608800 @default.
- W3092611426 doi "https://doi.org/10.1109/dac18072.2020.9218576" @default.
- W3092611426 hasPublicationYear "2020" @default.
- W3092611426 type Work @default.
- W3092611426 sameAs 3092611426 @default.
- W3092611426 citedByCount "8" @default.
- W3092611426 countsByYear W30926114262020 @default.
- W3092611426 countsByYear W30926114262021 @default.
- W3092611426 countsByYear W30926114262022 @default.
- W3092611426 countsByYear W30926114262023 @default.
- W3092611426 crossrefType "proceedings-article" @default.
- W3092611426 hasAuthorship W3092611426A5015236818 @default.
- W3092611426 hasAuthorship W3092611426A5038935739 @default.
- W3092611426 hasAuthorship W3092611426A5076019669 @default.
- W3092611426 hasAuthorship W3092611426A5076083402 @default.
- W3092611426 hasConcept C108583219 @default.
- W3092611426 hasConcept C11413529 @default.
- W3092611426 hasConcept C134306372 @default.
- W3092611426 hasConcept C154945302 @default.
- W3092611426 hasConcept C28855332 @default.
- W3092611426 hasConcept C33923547 @default.
- W3092611426 hasConcept C41008148 @default.
- W3092611426 hasConcept C45374587 @default.
- W3092611426 hasConcept C50644808 @default.
- W3092611426 hasConcept C73000952 @default.
- W3092611426 hasConceptScore W3092611426C108583219 @default.
- W3092611426 hasConceptScore W3092611426C11413529 @default.
- W3092611426 hasConceptScore W3092611426C134306372 @default.
- W3092611426 hasConceptScore W3092611426C154945302 @default.
- W3092611426 hasConceptScore W3092611426C28855332 @default.
- W3092611426 hasConceptScore W3092611426C33923547 @default.
- W3092611426 hasConceptScore W3092611426C41008148 @default.
- W3092611426 hasConceptScore W3092611426C45374587 @default.
- W3092611426 hasConceptScore W3092611426C50644808 @default.
- W3092611426 hasConceptScore W3092611426C73000952 @default.
- W3092611426 hasLocation W30926114261 @default.
- W3092611426 hasOpenAccess W3092611426 @default.
- W3092611426 hasPrimaryLocation W30926114261 @default.
- W3092611426 hasRelatedWork W2126887587 @default.
- W3092611426 hasRelatedWork W2354062721 @default.
- W3092611426 hasRelatedWork W2731899572 @default.
- W3092611426 hasRelatedWork W2939353110 @default.
- W3092611426 hasRelatedWork W2948658236 @default.
- W3092611426 hasRelatedWork W3003265541 @default.
- W3092611426 hasRelatedWork W3009238340 @default.
- W3092611426 hasRelatedWork W3215138031 @default.
- W3092611426 hasRelatedWork W4230611425 @default.
- W3092611426 hasRelatedWork W4287871663 @default.
- W3092611426 isParatext "false" @default.
- W3092611426 isRetracted "false" @default.
- W3092611426 magId "3092611426" @default.
- W3092611426 workType "article" @default.