Method
- Divide the available data into training and validation sets.
- Use a large number of hidden units.
- Use very small random initial values.
- Use a slow learning rate.
- Compute the validation error rate periodically during training.
- Stop training when the validation error rate "starts to go up".
It is crucial to realize that the validation error is not a good estimate of the generalization error. One method for getting an unbiased estimate of the generalization error is to run the net on a third set of data, the test set, that is not used at all during the training process. The error on the test set gives estimate on generalization; to have the outputs of the net approximate target values given inputs that are not in the training set.
Read more about this topic: Early Stopping
Famous quotes containing the word method:
“Traditional scientific method has always been at the very best 20-20 hindsight. It’s good for seeing where you’ve been. It’s good for testing the truth of what you think you know, but it can’t tell you where you ought to go.”
—Robert M. Pirsig (b. 1928)
“No method nor discipline can supersede the necessity of being forever on the alert. What is a course of history or philosophy, or poetry, no matter how well selected, or the best society, or the most admirable routine of life, compared with the discipline of looking always at what is to be seen? Will you be a reader, a student merely, or a seer? Read your fate, see what is before you, and walk on into futurity.”
—Henry David Thoreau (1817–1862)
“The country needs and, unless I mistake its temper, the country demands bold, persistent experimentation. It is common sense to take a method and try it. If it fails, admit it frankly and try another. But above all, try something. The millions who are in want will not stand idly by silently forever while the things to satisfy their needs are within easy reach.”
—Franklin D. Roosevelt (1882–1945)