首页> 外文期刊>Journal of complexity >Function approximation with zonal function networks with activation functions analogous to the rectified linear unit functions
【24h】

Function approximation with zonal function networks with activation functions analogous to the rectified linear unit functions

机译:使用具有类似于整流线性单位函数的激活函数的区域函数网络进行函数逼近

获取原文
获取原文并翻译 | 示例
           

摘要

A zonal function (ZF) network on the q dimensional sphere S-q is a network of the form x bar right arrow Sigma(n)(k=1)a(k)phi(x.x(k)) where phi : [-1, 1] - R is the activation function, x(k) is an element of S-q are the centers, and a(k) is an element of R. While the approximation properties of such networks are well studied in the context of positive definite activation functions, recent interest in deep and shallow networks motivate the study of activation functions of the form phi(t) = vertical bar t vertical bar, which are not positive definite. In this paper, we define an appropriate smoothness class and establish approximation properties of such networks for functions in this class. The centers can be chosen independently of the target function, and the coefficients are linear combinations of the training data. The constructions preserve rotational symmetries. (C) 2018 Elsevier Inc. All rights reserved.
机译:q维球面Sq上的区域功能(ZF)网络是形式为x条右箭头Sigma(n)(k = 1)a(k)phi(xx(k))的网络,其中phi:[-1, 1]-> R是激活函数,x(k)是Sq的元素是中心,而a(k)是R的元素。虽然在正定上下文中很好地研究了此类网络的逼近性质激活函数,近来对深层和浅层网络的关注激发了形式为phi(t)=垂直线t垂直线的激活函数的研究,这些函数不是正定的。在本文中,我们定义了一个合适的平滑度类,并为此类中的函数建立了此类网络的近似性质。可以独立于目标函数选择中心,并且系数是训练数据的线性组合。该构造保持旋转对称性。 (C)2018 Elsevier Inc.保留所有权利。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号