A Combination of Multi-state Activation Functions, Mean-normalisation and Singular Value Decomposition for learning Deep Neural Networks.
Chenghao CaiDengfeng KeYanyan XuKaile SuPublished in: IJCNN (2015)
Keyphrases
- singular value decomposition
- neural network
- activation function
- neural nets
- computer vision
- pattern recognition
- multilayer perceptron
- least squares
- feed forward
- genetic algorithm
- learning algorithm
- data sets
- state space
- singular values
- latent semantic indexing
- back propagation
- active learning
- artificial neural networks
- feature extraction
- image processing
- feature selection