Global Sparse Momentum SGD for Pruning Very Deep Neural Networks.
Xiaohan DingGuiguang DingXiangxin ZhouYuchen GuoJungong HanJi LiuPublished in: NeurIPS (2019)
Keyphrases
- neural network
- pattern recognition
- artificial neural networks
- genetic algorithm
- neural network model
- high dimensional
- fuzzy logic
- neural nets
- stochastic gradient descent
- multi layer
- avoid overfitting
- sparse matrix
- fuzzy systems
- convergence speed
- multilayer perceptron
- sparse coding
- feed forward
- convergence rate
- sparse representation
- back propagation
- linear combination