Sparse ternary connect: Convolutional neural networks using ternarized weights with enhanced sparsity.
Canran JinHeming SunShinji KimuraPublished in: ASP-DAC (2018)
Keyphrases
- convolutional neural networks
- sparse representation
- high dimensional
- sparsity constraints
- sparse approximation
- linear combination
- convolutional network
- sparse coding
- mixed norm
- weighted sum
- basis pursuit
- compressed sensing
- sparse approximations
- dictionary learning
- compressive sampling
- sparse data
- compressive sensing
- relative importance
- tensor factorization
- sparse reconstruction
- norm regularization
- regularized regression
- weighting scheme
- negative matrix factorization
- image classification
- signal processing
- denoising
- structured sparsity
- random projections
- sparsity inducing
- low rank
- neural network
- canonical correlation analysis
- weighted average
- sparse coefficients