SAI: Self-Adjusting Incremental Quantile Estimation for Sparse Training of Neural Networks on Hardware Accelerators.
Jianchao YangMei WenMinjin TangJunzhong ShenChunyuan ZhangPublished in: HPCC/DSS/SmartCity/DependSys (2021)
Keyphrases
- neural network
- training process
- training algorithm
- feedforward neural networks
- feed forward neural networks
- pattern recognition
- neural network training
- error back propagation
- online learning
- multi layer perceptron
- test set
- artificial neural networks
- high dimensional
- training patterns
- backpropagation algorithm
- reduced set
- multi layer
- feed forward
- fault diagnosis
- sparse representation
- avoid overfitting
- data driven
- incremental learning
- neural network structure
- neural network model
- training phase
- sparse data
- supervised learning
- radial basis function network
- training set
- batch mode
- incremental version
- multilayer neural network
- compressive sensing
- hidden layer
- neural nets
- self organizing maps
- back propagation
- training examples
- machine learning