论文信息 - A 0.3–2.6 TOPS/W precision-scalable processor for real-time large-scale ConvNets

A 0.3–2.6 TOPS/W precision-scalable processor for real-time large-scale ConvNets

A low-power precision-scalable processor for ConvNets or convolutional neural networks (CNN) is implemented in a 40nm technology. Its 256 parallel processing units achieve a peak 102GOPS running at 204MHz. To minimize energy consumption while maintaining throughput, this works is the first to both exploit the sparsity of convolutions and to implement dynamic precision-scalability enabling supply- and energy scaling. The processor is fully C-programmable, consumes 25-288mW at 204 MHz and scales efficiency from 0.3-2.6 real TOPS/W. This system hereby outperforms the state-of-the-art up to 3.9× in energy efficiency.

Marian Verhelst | Bert Moons | M. Verhelst | Bert Moons

[1] Vivienne Sze,et al. 14.5 Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks , 2016, ISSCC.

[2] Geoffrey E. Hinton,et al. ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[3] Yoshua Bengio,et al. Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[4] Joel Emer,et al. Eyeriss: an Energy-efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks Accessed Terms of Use , 2022 .

[5] Luca Benini,et al. Origami: A Convolutional Network Accelerator , 2015, ACM Great Lakes Symposium on VLSI.

[6] Marian Verhelst,et al. Energy-efficient ConvNets through approximate computing , 2016, 2016 IEEE Winter Conference on Applications of Computer Vision (WACV).