On the momentum term in gradient descent learning algorithms.
Ning QianPublished in: Neural Networks (1999)
Keyphrases
- learning algorithm
- learning rate
- error function
- update rule
- cost function
- machine learning algorithms
- machine learning
- training data
- active learning
- chan vese model
- convergence rate
- objective function
- training samples
- regularization term
- mumford shah functional
- back propagation
- loss function
- learning tasks
- learning problems
- reinforcement learning
- neural network
- training examples
- convergence speed
- model selection
- efficient learning
- decision tree learning