| 提高BP网络性能的一种方法 |
| |
| 引用本文： | 张宝金,胡青苗,曾梅光,田德新.提高BP网络性能的一种方法[J].东北大学学报(自然科学版),1999,20(1):3-110. |
| |
| 作者姓名： | 张宝金 胡青苗 曾梅光 田德新 |
| |
| 作者单位： | 东北大学理学院;武汉钢铁公司钢研所 |
| |
| 基金项目： | 国家科技攻关项目;95-527-01-02-02-03; |
| |
| 摘 要： | 分析了ＢＰ网络在目标量的极值区存在较大误差的原因,提出了目标量扩展的数据标度化方法·检验结果表明,该方法不但可以消除目标量极值区的大误差现象、提高网络训练的收敛速度,而且还可以使网络具有外推预测功能,即同时提高了ＢＰ网络的训练性能和应用性能·
|
| 关 键 词： | 人工神经网络 BP算法 标度化 网络性能 |
Method of Improving the Properties of BP Artificial Neural Networks |
| |
| Zhang Baojin,Hu Qingmiao,Zeng Meiguang,Tian Dexin.Method of Improving the Properties of BP Artificial Neural Networks[J].Journal of Northeastern University(Natural Science),1999,20(1):3-110. |
| |
| Authors: | Zhang Baojin Hu Qingmiao Zeng Meiguang Tian Dexin |
| |
| Abstract: | The reason that there existed big training errors in the extremum areas of the target parameters was analyzed and a new algorithm with target parameters value expansion in scale conversion was proposed. The results of training with the new algorithm showed that the big error phenomena was eliminated, and the rate of convergence of the network was increased. In addition, the network was endowed with extrapolation capability, which indicated that both properties of training and application were improved. |
| |
| Keywords: | artificial neural networks BP algorithm scale conversion properties of networks |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《东北大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《东北大学学报(自然科学版)》下载免费的PDF全文 |
