期刊
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
卷 36, 期 4, 页码 1639-1653出版社
SPRINGER BIRKHAUSER
DOI: 10.1007/s00034-016-0375-7
关键词
Artificial neural networks; Radial basis function; Gaussian kernel; Support vector machine; Euclidean distance; Cosine distance; Kernel fusion
In this paper, we propose a novel adaptive kernel for the radial basis function neural networks. The proposed kernel adaptively fuses the Euclidean and cosine distance measures to exploit the reciprocating properties of the two. The proposed framework dynamically adapts the weights of the participating kernels using the gradient descent method, thereby alleviating the need for predetermined weights. The proposed method is shown to outperform the manual fusion of the kernels on three major problems of estimation, namely nonlinear system identification, patter classification and function approximation.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据