On Selection of Training Data for Fast Learning of Neural Networks using Back Propagation

摘要

This study focuses on the subject of training data selection for neural networks using back propagation. Training data is analyzed and the data that effects the learning process is selected based on the idea of Critical points. The proposed method is applied to a classification problem where the task is to recognize the characters A,C from B,D. The experimental results show that the proposed method takes almost 1/7 of real and 1/12 of user training time required for the conventional method. The classification rate of the training and testing data are the same as it is with the conventional method.