Methods of NN Construction for the Problem of One-Variable Function Approximation

摘要：

Method of construction of fully connected twolayer neural network, which realize piecewisenonlinear approximation of a given one-variable junction y=f(x), is proposed The method involves determination of networks neurons number, which are necessary for function approximation with error not exceeding the given one, and determination of weights values for all neurons of network.

作者: Tomashevich N.S.

作者单位: Scientific Center of Neurocomputeis

会议类型: 国际会议

会议名称: 8th International Conference on Neural Information Processing(ICONIP 2001)(第八届国际神经信息处理大会)

会议地点: 上海

会议语种:英文

页码: 1527-1530

在线出版日期: 2001-11-14（万方平台首次上网日期，不代表论文的发表时间）