Issues
- It's not clear on how many cases to assign to the training and validation sets
- The result might highly depend on the algorithm which is used to split the data into training and validation set
- Notion of "increasing validation error" is ambiguous; it may go up and down numerous times during training. The safest approach is to train to convergence, then determine which iteration had the lowest validation error. This impairs fast training, one of the advantages of early stopping.
Read more about this topic: Early Stopping
Famous quotes containing the word issues:
“The hard truth is that what may be acceptable in elite culture may not be acceptable in mass culture, that tastes which pose only innocent ethical issues as the property of a minority become corrupting when they become more established. Taste is context, and the context has changed.”
—Susan Sontag (b. 1933)
“Your toddler will be “good” if he feels like doing what you happen to want him to do and does not happen to feel like doing anything you would dislike. With a little cleverness you can organize life as a whole, and issues in particular, so that you both want the same thing most of the time.”
—Penelope Leach (20th century)
“To make life more bearable and pleasant for everybody, choose the issues that are significant enough to fight over, and ignore or use distraction for those you can let slide that day. Picking your battles will eliminate a number of conflicts, and yet will still leave you feeling in control.”
—Lawrence Balter (20th century)