A Hidden Feature Selection Method based on l2, 0-Norm Regularization for Training Single-hidden-layer Neural Networks.
Zhiwei LiuYuanlong YuZhenzhen SunPublished in: SSCI (2019)
Keyphrases
- neural network
- training process
- feature selection
- feed forward neural networks
- feedforward neural networks
- cost function
- support vector machine
- training algorithm
- feature set
- training samples
- probabilistic model
- missing data
- bp neural network
- text categorization
- neural nets
- bayesian inference
- stochastic gradient descent
- extreme learning machine
- recurrent neural networks
- bayesian framework
- support vector machine svm
- supervised learning
- multi class
- training set
- support vector
- training data
- genetic algorithm