摘要： 提出一种基于进化算法的ＳＦＳＮ（Ｓａｍｐｌｉｎｇ Ｆｒｅｑｕｅｎｃｙ－ｓｅｎｓｉｔｉｖｅ Ｎｅｔｗｏｒｋ）神经网络分类量化方法．该算法把并行全局寻优的进化算法ＥＰ（Ｅｖｏｌｕｔｉｏｎａｒｙ Ｐｒｏｇｒａｍｍｉｎｇ）融合进ＳＦＳＮ神经网络．使神经网络的结构优化与参数优化同时完成．即同时解决了最佳分类数与最佳矢量量化问题．实验结果证实了算法的有效性．

关键词: 神经网络, 进化算法, 矢量量化, 聚类

Abstract: In this paper. an evolutionary proaramming-based sampling frequency-sensitive networks is proposed. The algorithm effectively fuses the powerful parallel search optimization tool EP(evolutionary programming) and SFSN neural network. The proposed method has an advantage that the optimal solution of architecture and parameters can be get simultaneously. So the classification network can get the optimal number of clusters and the optimal vector quantization. The results of the experiment are given to prove it.

中图分类号: 

• TP183