Joint Pruning & Quantization for Extremely Sparse Neural Networks.
Po-Hsiang YuSih-Sian WuJan P. KloppLiang-Gee ChenShao-Yi ChienPublished in: CoRR (2020)
Keyphrases
- neural network
- pattern recognition
- feed forward
- neural network model
- fuzzy logic
- pruning algorithms
- genetic algorithm
- multilayer perceptron
- high dimensional
- artificial neural networks
- search space
- avoid overfitting
- pruning algorithm
- sparse representation
- effective pruning
- data sets
- compressed sensing
- sparse data
- random projections
- activation function
- recurrent neural networks
- back propagation
- computational complexity
- fuzzy systems
- multi layer
- rule extraction
- competitive learning
- quantization error
- joint optimization
- radial basis function
- image compression
- tree pruning