期刊
NEURAL PROCESSING LETTERS
卷 28, 期 2, 页码 63-79出版社
SPRINGER
DOI: 10.1007/s11063-008-9082-0
关键词
activation functions; dynamical system identification; Levenberg-Marquardt algorithm
资金
- TOBB Economics and Technology University [ETU BAP-2006/04]
- TUBITAK [107E137]
Feedforward neural network structures have extensively been considered in the literature. In a significant volume of research and development studies hyperbolic tangent type of a neuronal nonlinearity has been utilized. This paper dwells on the widely used neuronal activation functions as well as two new ones composed of sines and cosines, and a sinc function characterizing the firing of a neuron. The viewpoint here is to consider the hidden layer(s) as transforming blocks composed of nonlinear basis functions, which may assume different forms. This paper considers 8 different activation functions which are differentiable and utilizes Levenberg-Marquardt algorithm for parameter tuning purposes. The studies carried out have a guiding quality based on empirical results on several training data sets.
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