SeFAct: selective feature activation and early classification for CNNs-猫眼课题宝

DOI：

10.1145/3287624.3287663

发表时间：

2019-01

期刊：

Proceedings of the 24th Asia and South Pacific Design Automation Conference

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

F. S. Snigdha;Ibrahim Ahmed;Susmita Dey Manasi;Meghna G. Mankalale;Jiang Hu;S. Sapatnekar

提出了一种用于卷积神经网络(CNN)硬件加速器的动态能量降低方法。使用了两种方法：(1)自适应数据依赖方案，通过缩小可能激活的类来选择性地激活所有神经元的子集(2)静态位宽减少。前者适用于CNN的后期，而后者则更有效地应用于早期。即使考虑实施开销，结果也显示节能20%-25%，但精确度损失5%-10%。

This work presents a dynamic energy reduction approach for hardware accelerators for convolutional neural networks (CNN). Two methods are used: (1) an adaptive data-dependent scheme to selectively activate a subset of all neurons, by narrowing down the possible activated classes (2) static bitwidth reduction. The former is applied in late layers of the CNN, while the latter is more effective in early layers. Even accounting for the implementation overheads, the results show 20%--25% energy savings with 5--10% accuracy loss.

SeFAct: selective feature activation and early classification for CNNs

文献摘要

相似文献