Fast unit pruning algorithm for feedforward neural network design

被引：30

作者：

Qiao, Jun-fei ^{[1
]}

Zhang, Ying ^{[1
]}

Han, Hong-gui ^{[1
]}

机构：

[1] Beijing Univ Technol, Coll Elect & Control Engn, Beijing 100022, Peoples R China

来源：

APPLIED MATHEMATICS AND COMPUTATION | 2008年 / 205卷 / 02期

基金：

中国国家自然科学基金;

关键词：

D O I：

10.1016/j.amc.2008.05.049

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

A fast unit pruning algorithm for feedforward neural network is presented and the way used by the algorithm which based on optimal brain surgeon (OBS) is to remove the unneeded hidden units directly so that carry out the self-organization design on the architecture of neural networks. The algorithm is tested on several modeling problems, and is compared with OBS. It is found that the fast unit pruning algorithm is much more efficient than OBS which can not only reduce the complexity of the network but also accelerate the learning speed. (C) 2008 Elsevier Inc. All rights reserved.

引用

页码：622 / 627

页数：6

共 9 条

[1]

[Anonymous], 1990, ADV NEURAL INFORM PR

[2]

[Anonymous], NEURAL NETWORKS

[3]

Hassibi B., 1993, IEEE INT C NEUR NETW, V1-3, P293, DOI [10.1109/ICNN.1993.298572, DOI 10.1109/ICNN.1993.298572]

[4]

LO JT, 1999, NEURAL NETWORKS, V3, P1678

[5]

MENG J, 2005, TOOLS ARTIFICIAL INT, P5

[6]

MESSER K, 2000, PATT REC 2000 P 15 I, V2, P981

[7]

MESSER K, 1998, P 9 BRIT MACH VIS C, P235

[8]

QIAN LI, 2005, TSINGHUA U SCI TECHN, V45, P831

[9]

SHUDONG HE, 1998, CONTROL THEORY APPL, V15, P313

← 1 →