Research Article
Efficient Architecture for Convolution and Softmax Function in Deep Learning Accelerator
@INPROCEEDINGS{10.1007/978-3-030-67720-6_43, author={Zhenyu Jiang and Zhifeng Zhang and Haoqi Ren and Jun Wu}, title={Efficient Architecture for Convolution and Softmax Function in Deep Learning Accelerator}, proceedings={Communications and Networking. 15th EAI International Conference, ChinaCom 2020, Shanghai, China, November 20-21, 2020, Proceedings}, proceedings_a={CHINACOM}, year={2021}, month={2}, keywords={Convolutional neural network Hardware architecture Convolution Winograd algorithm Softmax function}, doi={10.1007/978-3-030-67720-6_43} }- Zhenyu Jiang
Zhifeng Zhang
Haoqi Ren
Jun Wu
Year: 2021
Efficient Architecture for Convolution and Softmax Function in Deep Learning Accelerator
CHINACOM
Springer
DOI: 10.1007/978-3-030-67720-6_43
Abstract
Convolutional neural network (CNN) has been widely used in deep learning. However, thehardwareconsumptionof the convolutional neural networkis very large. Traditional Central Processing Units (CPUs) and Graphic Processing Units (GPUs) are inefficient and expensive for neural network, so an efficient hardware design is required. The proposed design based on Digital Signal Processor (DSP) has rapid operating speed and strong computation ability for training and inference of CNN. In this paper, the hardware architecture of convolution and softmax function is specially optimized. Winograd algorithm can reduce multiplications of convolution, thus decreases hardware complexity, since multiplication is much more complex in hardware implementation than addition. The softmax function is also simplified by replacing divider by subtractor and logarithmic function which cost fewer resources. The proposed hardware architecture dramatically decreases the complexity and hardware resources.
