Training neural networks to be insensitive to weight random variations.
Massimo ContiSimone OrcioniClaudio TurchettiPublished in: Neural Networks (2000)
Keyphrases
- neural network
- training process
- training algorithm
- feedforward neural networks
- feed forward neural networks
- pattern recognition
- back propagation
- neural network training
- backpropagation algorithm
- weight update
- fuzzy logic
- multi layer
- activation function
- recurrent networks
- error back propagation
- neural network structure
- training phase
- real time
- weighted sums
- neural network model
- fault diagnosis
- test set
- weighting scheme
- neural nets
- training set
- training patterns
- training samples
- genetic algorithm
- computer vision
- case study
- training data
- artificial neural networks
- active learning
- online learning
- rule extraction
- uniformly distributed
- multilayer perceptron
- recurrent neural networks