### Abstract

Using backpropagation algorithm (BP) to train neural networks is a widely adopted practice in both theory and practical applications. But its distributed weight representation, that is the weight matrix of final network after training by using BP are usually not sparsified, and prohibits its use in the rule discovery of inherent functional relations between the input and output data, so in this aspect some kinds of structure optimization are needed to improve its poor performance. In this paper with this in mind a new method to prune neural networks is proposed based on some statistical quantities of neural networks. Comparing with the other known pruning methods such as structural learning with forgetting (SLF) and RPROP algorithm, the proposed method can attain comparable or even better results over these methods without evident increase of the computational load. Detailed simulations using the Iris data sets exhibit our above assertion.

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

Title of host publication | Proceedings of the International Joint Conference on Neural Networks |

Place of Publication | Piscataway, NJ, United States |

Publisher | IEEE |

Pages | 449-454 |

Number of pages | 6 |

Volume | 6 |

Publication status | Published - 2000 |

Externally published | Yes |

Event | International Joint Conference on Neural Networks (IJCNN'2000) - Como, Italy Duration: 2000 Jul 24 → 2000 Jul 27 |

### Other

Other | International Joint Conference on Neural Networks (IJCNN'2000) |
---|---|

City | Como, Italy |

Period | 00/7/24 → 00/7/27 |

### Fingerprint

### ASJC Scopus subject areas

- Software

### Cite this

*Proceedings of the International Joint Conference on Neural Networks*(Vol. 6, pp. 449-454). Piscataway, NJ, United States: IEEE.

**New method to prune the neural network.** / Wan, Weishui; Hirasawa, Kotaro; Furuzuki, Takayuki; Jin, Chunzhi.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*Proceedings of the International Joint Conference on Neural Networks.*vol. 6, IEEE, Piscataway, NJ, United States, pp. 449-454, International Joint Conference on Neural Networks (IJCNN'2000), Como, Italy, 00/7/24.

}

TY - GEN

T1 - New method to prune the neural network

AU - Wan, Weishui

AU - Hirasawa, Kotaro

AU - Furuzuki, Takayuki

AU - Jin, Chunzhi

PY - 2000

Y1 - 2000

N2 - Using backpropagation algorithm (BP) to train neural networks is a widely adopted practice in both theory and practical applications. But its distributed weight representation, that is the weight matrix of final network after training by using BP are usually not sparsified, and prohibits its use in the rule discovery of inherent functional relations between the input and output data, so in this aspect some kinds of structure optimization are needed to improve its poor performance. In this paper with this in mind a new method to prune neural networks is proposed based on some statistical quantities of neural networks. Comparing with the other known pruning methods such as structural learning with forgetting (SLF) and RPROP algorithm, the proposed method can attain comparable or even better results over these methods without evident increase of the computational load. Detailed simulations using the Iris data sets exhibit our above assertion.

AB - Using backpropagation algorithm (BP) to train neural networks is a widely adopted practice in both theory and practical applications. But its distributed weight representation, that is the weight matrix of final network after training by using BP are usually not sparsified, and prohibits its use in the rule discovery of inherent functional relations between the input and output data, so in this aspect some kinds of structure optimization are needed to improve its poor performance. In this paper with this in mind a new method to prune neural networks is proposed based on some statistical quantities of neural networks. Comparing with the other known pruning methods such as structural learning with forgetting (SLF) and RPROP algorithm, the proposed method can attain comparable or even better results over these methods without evident increase of the computational load. Detailed simulations using the Iris data sets exhibit our above assertion.

UR - http://www.scopus.com/inward/record.url?scp=0033703422&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033703422&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0033703422

VL - 6

SP - 449

EP - 454

BT - Proceedings of the International Joint Conference on Neural Networks

PB - IEEE

CY - Piscataway, NJ, United States

ER -