Minimizing the Quadratic Training Error of a Sigmoid Neuron Is Hard.
Jirí SímaPublished in: ALT (2001)
Keyphrases
- training error
- error rate
- generalization error
- hidden layer
- activation function
- classification error
- adaboost algorithm
- neural network
- gradient method
- computational complexity
- base classifiers
- prediction error
- cross validation
- multiclass classification
- objective function
- training set
- boosting algorithms
- weak classifiers
- binary classification
- back propagation
- artificial neural networks
- feed forward