Available in: Deep Learning
When adaptive learning rate is disabled, the magnitude of the weight updates are determined by the user specified learning rate (potentially annealed), and are a function of the difference between the predicted value and the target value. That difference, generally called delta, is only available at the output layer. To correct the output at each hidden layer, back propagation is used. Momentum modifies back propagation by allowing prior iterations to influence the current update. Using the momentum parameter can aid in avoiding local minima and the associated instability. Too much momentum can lead to instabilities. That’s why the momentum is best ramped up slowly.
This parameter is only active when adaptive learning rate is disabled.