Neural network training algorithm that can predict generalization capacity
Jin Lu, Wenli Xu, Han Zengjin · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1995
Neural network training requires a large quantity of samples and consumes a great deal of computing time. Despite this, we still do not know the generalization capacity of a trained neural network in a certain domain. In this paper, we propose an algorithm for training neural networks to approximate polynomials. This algorithm can work with a relatively small sample set and predict the generalization capacity of the learned neural network. Simulation results demonstrate the property of this algorithm.