Reducing the Training Times of Neural Classifiers with Dataset Condensing

摘要

In this paper we apply a k-nearest-neighbour-based data condensing algorithm to the training sets of multi-layer perceptron neural networks. By removing the overlapping data and retaining only training exemplars adjacent to the decision boundary we are able to significantly speed the network training time while achieving an undegraded misclassification rate compared to a network trained on the unedited training set. We report results on a range of synthetic and real datasets which indicate that a speed-up of an order of magnitude in the network training time is typical.