Abstract
In this paper, we propose a novel adaptive kernel for the radial basis function (RBF) 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, pattern classification and function approximation.
Abstract (translated)
本文提出了一种新的径向基函数(RBF)神经网络自适应核。该核自适应地融合了欧几里得距离和余弦距离两种度量,利用了两者的往复特性。该框架采用梯度下降法动态调整参与核的权值,从而减少了对预定权值的需要。结果表明,该方法在非线性系统辨识、模式分类和函数逼近三个主要估计问题上优于人工核融合。
URL
https://arxiv.org/abs/1905.03546