A geometric method to obtain error-correcting classification by neural networks with fewer hidden units.
Vera KurkováPaul C. KainenPublished in: ICNN (1996)
Keyphrases
- neural network
- support vector machine svm
- classification accuracy
- classification algorithm
- feed forward neural networks
- pattern recognition
- text classification
- model selection
- error correcting
- feature extraction
- nonlinear functions
- training phase
- data sets
- training set
- training samples
- training process
- active learning
- neural nets
- class labels
- training algorithm
- optimization method
- support vector
- feature space
- input data
- feature set
- dimensionality reduction
- supervised learning
- support vector machine