SME: ReRAM-based Sparse-Multiplication-Engine to Squeeze-Out Bit Sparsity of Neural Network.
Fangxin LiuWenbo ZhaoZhezhi HeZongwu WangYilong ZhaoTao YangJingnai FengXiaoyao LiangLi JiangPublished in: ICCD (2021)
Keyphrases
- neural network
- sparse representation
- high dimensional
- sparsity constraints
- sparse approximation
- compressive sampling
- compressive sensing
- artificial neural networks
- integer arithmetic
- dictionary learning
- sparse coding
- mixed norm
- basis pursuit
- floating point
- tensor factorization
- sparse reconstruction
- fuzzy logic
- back propagation
- group lasso
- sparse data
- neural network model
- neural network is trained
- norm regularization
- neural nets
- fault diagnosis
- compressed sensing
- linear combination
- sparse approximations
- regularized regression
- canonical correlation analysis
- genetic algorithm
- fuzzy neural network
- face recognition
- denoising
- low rank
- variable selection
- multi layer
- s box
- bp neural network
- multilayer perceptron
- structured sparsity
- feed forward
- radial basis function
- self organizing maps
- high dimensional data
- image representation
- sparse coefficients
- image classification
- network architecture
- negative matrix factorization