Appeared in Neural Networks 1998 Automatic Early Stopping Using Cross Validation: Quantifying the Criteria

Cross validation can be used to detect when over tting starts during supervised training of a neural network; training is then stopped before convergence to avoid the overtting (\early stopping"). The exact criterion used for cross validation based early stopping, however, is chosen in an ad-hoc fashion by most researchers or training is stopped interactively. To aid a more well-founded selection of the stopping criterion, 14 di erent automatic stopping criteria from 3 classes were evaluated empirically for their e ciency and e ectiveness in 12 di erent classi cation and approximation tasks using multi layer perceptrons with RPROP training. The experiments show that on the average slower stopping criteria allow for small improvements in generalization (on the order of 4%), but cost about factor 4 longer training time.