### Abstract

A traditional Radial basis function (RBF) network takes Gaussian functions as its basis functions and adopts the least squares(LS) criterion as the objective function. However, it is difficult to use Gaussian functions to approximate constant values. If a function has nearly constant values in some intervals, the RBF network will be found inefficient in approximating these values. In this paper an RBF network which uses composite of sigmoidal functions to replace the Gaussian functions as the basis function of the network is proposed. It is also illustrated that the shape of the activation function can be constructed to be a similar rectangular or Gaussian function. Thus, the constant-valued functions can be approximated accurately by an RBF network. A robust objective function is also adopted in the network to replace the LS objective function. Experimental results demonstrated that the proposed network has better capability of approximation to underlying functions with a fast learning speed and high robustness to outliers.

Original language | English |
---|---|

Pages | 496-501 |

Number of pages | 6 |

Publication status | Published - 1996 Jan 1 |

Event | Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4) - Washington, DC, USA Duration: 1996 Jun 3 → 1996 Jun 6 |

### Other

Other | Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4) |
---|---|

City | Washington, DC, USA |

Period | 96-06-03 → 96-06-06 |

### Fingerprint

### All Science Journal Classification (ASJC) codes

- Software

### Cite this

*Sigmoidal radial basis function neural network for function approximation*. 496-501. Paper presented at Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4), Washington, DC, USA, .

}

**Sigmoidal radial basis function neural network for function approximation.** / Tsai, Jea Rong; Chung, Pau-Choo; Chang, Chein I.

Research output: Contribution to conference › Paper

TY - CONF

T1 - Sigmoidal radial basis function neural network for function approximation

AU - Tsai, Jea Rong

AU - Chung, Pau-Choo

AU - Chang, Chein I.

PY - 1996/1/1

Y1 - 1996/1/1

N2 - A traditional Radial basis function (RBF) network takes Gaussian functions as its basis functions and adopts the least squares(LS) criterion as the objective function. However, it is difficult to use Gaussian functions to approximate constant values. If a function has nearly constant values in some intervals, the RBF network will be found inefficient in approximating these values. In this paper an RBF network which uses composite of sigmoidal functions to replace the Gaussian functions as the basis function of the network is proposed. It is also illustrated that the shape of the activation function can be constructed to be a similar rectangular or Gaussian function. Thus, the constant-valued functions can be approximated accurately by an RBF network. A robust objective function is also adopted in the network to replace the LS objective function. Experimental results demonstrated that the proposed network has better capability of approximation to underlying functions with a fast learning speed and high robustness to outliers.

AB - A traditional Radial basis function (RBF) network takes Gaussian functions as its basis functions and adopts the least squares(LS) criterion as the objective function. However, it is difficult to use Gaussian functions to approximate constant values. If a function has nearly constant values in some intervals, the RBF network will be found inefficient in approximating these values. In this paper an RBF network which uses composite of sigmoidal functions to replace the Gaussian functions as the basis function of the network is proposed. It is also illustrated that the shape of the activation function can be constructed to be a similar rectangular or Gaussian function. Thus, the constant-valued functions can be approximated accurately by an RBF network. A robust objective function is also adopted in the network to replace the LS objective function. Experimental results demonstrated that the proposed network has better capability of approximation to underlying functions with a fast learning speed and high robustness to outliers.

UR - http://www.scopus.com/inward/record.url?scp=0029748120&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029748120&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0029748120

SP - 496

EP - 501

ER -