ZeNA: Zero-Aware Neural Network Accelerator.

IEEE Design & Test(2018)

引用 78|浏览110
暂无评分
摘要
It has been observed that the majority of the kernel weights and input activations in the state-of-the-art convolution neural networks (CNNs) have zero values. This article proposes a CNN hardware accelerator that exploits this property to achieve significant performance and energy improvements.
更多
查看译文
关键词
Convolutional neural networks,Random access memory,Machine learning,Computer architecture,Neural networks,Resource management
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要