SPARK: Scalable and Precision-Aware Acceleration of Neural Networks via Efficient Encoding.
Fangxin LiuNing YangHaomin LiZongwu WangZhuoran SongSongwen PeiLi JiangPublished in: HPCA (2024)
Keyphrases
- neural network
- highly scalable
- computationally expensive
- lightweight
- cost effective
- data sets
- bit wise
- fuzzy systems
- high precision
- back propagation
- computationally efficient
- artificial neural networks
- video sequences
- data structure
- multiscale
- training data
- database systems
- information systems
- machine learning
- high recall
- database