STICKER: An Energy-Efficient Multi-Sparsity Compatible Accelerator for Convolutional Neural Networks in 65-nm CMOS.
Zhe YuanYongpan LiuJinshan YueYixiong YangJingyu WangXiaoyu FengJian ZhaoXueqing LiHuazhong YangPublished in: IEEE J. Solid State Circuits (2020)
Keyphrases
- convolutional neural networks
- convolutional network
- high dimensional
- silicon on insulator
- high speed
- energy efficient
- cmos technology
- sensor networks
- wireless sensor networks
- sparse representation
- low power
- power supply
- nm technology
- energy consumption
- power consumption
- parallel implementation
- energy saving
- analog vlsi
- compute intensive
- real time