Lance: efficient low-precision quantized winograd convolution for neural networks based on graphics processing units

Accelerating deep convolutional neural networks has become an active topic and sparked an interest in academia and industry. In this paper, we propose an efficient low-precision quantized Winograd convolution algorithm, called LANCE, which combines the advantages of fast convolution and quantization...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings of the ... IEEE International Conference on Acoustics, Speech and Signal Processing (1998) s. 3842 - 3846
Hlavní autoři: Li, Guangli, Liu, Lei, Wang, Xueying, Ma, Xiu, Feng, Xiaobing
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.05.2020
Témata:
Convolution
deep learning
Graphics processing units
Industries
linear quantization
low-precision computing
Neural networks
Quantization (signal)
Signal processing algorithms
Speech processing
Winograd convolution
ISSN:2379-190X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Accelerating deep convolutional neural networks has become an active topic and sparked an interest in academia and industry. In this paper, we propose an efficient low-precision quantized Winograd convolution algorithm, called LANCE, which combines the advantages of fast convolution and quantization techniques. By embedding linear quantization operations into the Winograd-domain, the fast convolution can be performed efficiently under low-precision computation on graphics processing units. We test neural network models with LANCE on representative image classification datasets, including SVHN, CIFAR, and ImageNet. The experimental results show that our 8-bit quantized Winograd convolution improves the performance by up to 2.40× over the full-precision convolution with trivial accuracy loss.
ISSN:2379-190X
DOI:10.1109/ICASSP40776.2020.9054562