Matches in SemOpenAlex for { <https://semopenalex.org/work/W4207029376> ?p ?o ?g. }
Showing items 1 to 64 of
64
with 100 items per page.
- W4207029376 abstract "State-of-the-art Convolutional Neural Networks (CNNs) have become increasingly accurate. However, hundreds or thousands of megabytes data are involved to store them, making these networks also computationally expensive. For certain applications, such as Internet-of-Things (IoT), where such CNNs are to be implemented on resource-constrained and memory-constrained platforms, including Field-Programmable Gate Arrays (FPGAs) and embedded devices, CNN architectures and parameters have to be small and efficient. In this paper, an evolutionary algorithm (EA) based adaptive integer quantisation method is proposed to reduce network size. The proposed method uses single objective rank-based evolutionary strategy to find the best quantisation bin boundary for fixed quantised bit width. The performance of the proposed method is evaluated on a small CNN, the LeNet-5 architecture, using the CIFAR-10 dataset. The aim is to devise a methodology that allows adaptive quantisation of both weights and bias from 32-bit floating point to 8-bit integer representation for LeNet-5, while retaining accuracy. The experiments compare straight-forward (linear) quantisation from 32-bits to 8-bits with the proposed adaptive quantisation method. The results show that the proposed method is capable of quantising CNNs to lower bit width representation with only a slight loss in classification accuracy." @default.
- W4207029376 created "2022-01-26" @default.
- W4207029376 creator A5034601593 @default.
- W4207029376 creator A5045134787 @default.
- W4207029376 creator A5048789717 @default.
- W4207029376 creator A5056241158 @default.
- W4207029376 date "2021-12-05" @default.
- W4207029376 modified "2023-09-27" @default.
- W4207029376 title "Adaptive Integer Quantisation for Convolutional Neural Networks through Evolutionary Algorithms" @default.
- W4207029376 doi "https://doi.org/10.1109/ssci50451.2021.9659887" @default.
- W4207029376 hasPublicationYear "2021" @default.
- W4207029376 type Work @default.
- W4207029376 citedByCount "0" @default.
- W4207029376 crossrefType "proceedings-article" @default.
- W4207029376 hasAuthorship W4207029376A5034601593 @default.
- W4207029376 hasAuthorship W4207029376A5045134787 @default.
- W4207029376 hasAuthorship W4207029376A5048789717 @default.
- W4207029376 hasAuthorship W4207029376A5056241158 @default.
- W4207029376 hasConcept C11413529 @default.
- W4207029376 hasConcept C154945302 @default.
- W4207029376 hasConcept C17744445 @default.
- W4207029376 hasConcept C199360897 @default.
- W4207029376 hasConcept C199539241 @default.
- W4207029376 hasConcept C2776359362 @default.
- W4207029376 hasConcept C28855332 @default.
- W4207029376 hasConcept C41008148 @default.
- W4207029376 hasConcept C42935608 @default.
- W4207029376 hasConcept C50644808 @default.
- W4207029376 hasConcept C81363708 @default.
- W4207029376 hasConcept C84211073 @default.
- W4207029376 hasConcept C9390403 @default.
- W4207029376 hasConcept C94625758 @default.
- W4207029376 hasConcept C97137487 @default.
- W4207029376 hasConceptScore W4207029376C11413529 @default.
- W4207029376 hasConceptScore W4207029376C154945302 @default.
- W4207029376 hasConceptScore W4207029376C17744445 @default.
- W4207029376 hasConceptScore W4207029376C199360897 @default.
- W4207029376 hasConceptScore W4207029376C199539241 @default.
- W4207029376 hasConceptScore W4207029376C2776359362 @default.
- W4207029376 hasConceptScore W4207029376C28855332 @default.
- W4207029376 hasConceptScore W4207029376C41008148 @default.
- W4207029376 hasConceptScore W4207029376C42935608 @default.
- W4207029376 hasConceptScore W4207029376C50644808 @default.
- W4207029376 hasConceptScore W4207029376C81363708 @default.
- W4207029376 hasConceptScore W4207029376C84211073 @default.
- W4207029376 hasConceptScore W4207029376C9390403 @default.
- W4207029376 hasConceptScore W4207029376C94625758 @default.
- W4207029376 hasConceptScore W4207029376C97137487 @default.
- W4207029376 hasLocation W42070293761 @default.
- W4207029376 hasOpenAccess W4207029376 @default.
- W4207029376 hasPrimaryLocation W42070293761 @default.
- W4207029376 hasRelatedWork W2166734522 @default.
- W4207029376 hasRelatedWork W2219661328 @default.
- W4207029376 hasRelatedWork W2276486856 @default.
- W4207029376 hasRelatedWork W2376815915 @default.
- W4207029376 hasRelatedWork W2786216825 @default.
- W4207029376 hasRelatedWork W2794365619 @default.
- W4207029376 hasRelatedWork W2892054964 @default.
- W4207029376 hasRelatedWork W3161084567 @default.
- W4207029376 hasRelatedWork W3216547854 @default.
- W4207029376 hasRelatedWork W2182612463 @default.
- W4207029376 isParatext "false" @default.
- W4207029376 isRetracted "false" @default.
- W4207029376 workType "article" @default.