Using separable functional network for function approximation
Yongquan Zhou, Bai Liu, Huajuan Huang, Xingqong Wei · 2008
In this paper, separable functional network architecture and a learning algorithm of separable functional network are proposed, the learning of functional parameters use Lagrange multipliers by means of auxiliary function and solving a system of linear equations obtain parameters. An experiment in approximating typical continuous functions is given. The results show that the learning algorithm presented in the paper has excellent performance in approximation error.