Enabling High-Efficient ReRAM-Based CNN Training Via Exploiting Crossbar-Level Insignificant Writing Elimination.
Lening WangQiyu WanPeixun MaJing WangMinsong ChenShuaiwen Leon SongXin FuPublished in: IEEE Trans. Computers (2023)
Keyphrases
- high efficiency
- neural network
- levels of abstraction
- cost effective
- artificial neural networks
- test set
- training samples
- low memory requirements
- training process
- high precision
- training examples
- object detection
- training set
- learning environment
- image processing
- feature selection
- search engine
- artificial intelligence