摘要
为了提高神经网络求解问题的性能,有必要将多个网络融合。该文提出一种多网络线性融合的方法———基于最小均方差的遗传算法方法。使用此方法能显著提高网络的性能。仿真结果证明了该方法的有效性,具有实际应用价值。
Multi-Network fusion is useful to improve the performance of neural networks.In this paper,a linear fusion approach -genetic approach based on minimal mean squared error,is presented,which can greatly improve the performance of neural networks.Simulation result demonstrates the effectiveness of the proposed approach,which has great application values.
出处
《计算机工程与应用》
CSCD
北大核心
2003年第21期87-89,共3页
Computer Engineering and Applications
