A method to resolve the overfitting problem in recurrent neural networks for prediction of complex systems’ behavior
A method to resolve the overfitting problem in recurrent neural networks for prediction of complex systems’ behavior
复制标题
一种解决递归神经网络中过度拟合问题的方法，用于预测复杂系统的行为
DOI：
10.1109/ijcnn.2008.4634332
复制
发表时间：
2008
期刊：
影响因子：
--
通讯作者：
Mohammad Saraee
中科院分区：
文献类型：
--
作者：
K. Mahdaviani;Helga Mazyar;Saeed Majidi;Mohammad Saraee
本文提出了一种新的方法来解决预测复杂系统空间行为的过拟合问题。当神经网络失去其泛化能力时，就会出现这个问题。该方法基于递归神经网络的训练，并利用模拟退火法对其泛化进行优化。本文的主要工作是基于集成神经网络的思想。最后给出了该方法在两个样本数据集上的应用结果，说明了该方法的有效性。
In this paper a new method to resolve the overfitting problem for predicting complex systemspsila behavior has been proposed. This problem occurs when a neural network loses its generalization. The method is based on the training of recurrent neural networks and using simulated annealing for the optimization of their generalization. The major work is done based on the idea of ensemble neural networks. Finally the results of using this method on two sample datasets are presented and the effectiveness of this method is illustrated.