%0 Conference Proceedings
%T Accelerating Large-Scale Deep Convolutional Neural Networks on Multi-core Vector Accelerators
%+ College of Computer Science [Changsha]
%A Liu, Zhong
%A Ma, Sheng
%A Li, Cheng
%A Chen, Haiyan
%Z Part 2: AI
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B 17th IFIP International Conference on Network and Parallel Computing (NPC)
%C Zhengzhou, China
%Y Xin He
%Y En Shao
%Y Guangming Tan
%I Springer International Publishing
%3 Network and Parallel Computing
%V LNCS-12639
%P 68-79
%8 2020-09-28
%D 2020
%R 10.1007/978-3-030-79478-1_6
%K Multi-core vector accelerators
%K Convolutional neural network
%K Vectorization
%K AlexNet
%K VGG
%K GoogleNet
%K ResNet
%Z Computer Science [cs]Conference papers
%X This paper proposes an efficient algorithm mapping method for accelerating deep convolutional neural networks, which includes: (1) Proposing an efficient transformation method, which converts CNN’s convolutional layer and fully connected layer computations into efficient large-scale matrix multiplication computations, and converts pooling layer computations into efficient matrix row computations; (2) Designing a set of general and efficient vectorization method for convolutional layer, fully connected layer and pooling layer on the vector accelerator. The experimental results on the accelerator show that the average computing efficiency of convolution layer and full connected layer of AlexNet, VGG-19, GoogleNet and ResNet-50 are 93.3% and 93.4% respectively, and the average data access efficiency of pooling layer is 70%.
%G English
%Z TC 10
%Z WG 10.3
%2 https://inria.hal.science/hal-03768763/document
%2 https://inria.hal.science/hal-03768763/file/511910_1_En_6_Chapter.pdf
%L hal-03768763
%U https://inria.hal.science/hal-03768763
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-TC
%~ IFIP-TC10
%~ IFIP-NPC
%~ IFIP-WG10-3
%~ IFIP-LNCS-12639