| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 4628450 | Applied Mathematics and Computation | 2013 | 7 Pages |
Abstract
In this paper, the approximation capability of feed-forward neural networks (FNNs) formed as ∑k=1nckϕ(x-xk) is considered, where ϕ is an even periodic continuous function and xkxk’s are scattered data. The best approximation error of trigonometric polynomials and the continuous modulus of the activation function ϕ are introduced to describe the approximation capability of FNNs.
Related Topics
Physical Sciences and Engineering
Mathematics
Applied Mathematics
Authors
Shaobo Lin, Xiaofei Guo, Feilong Cao, Zongben Xu,