| 零误差密度函数准则的BP神经网络学习研究 |
| |
| 引用本文： | 邹修明,杨赛,孙怀江.零误差密度函数准则的BP神经网络学习研究[J].淮阴师范学院学报(自然科学版),2010,9(4). |
| |
| 作者姓名： | 邹修明 杨赛 孙怀江 |
| |
| 作者单位： | 1. 淮阴师范学院,物理与电子电气工程学院,江苏,淮安,223300;南京理工大学,计算机科学与技术学院,江苏,南京,210094
2. 南京理工大学,计算机科学与技术学院,江苏,南京,210094 |
| |
| 摘 要： | BP神经网络的学习通常以均方误差函数(MSE)为目标函数,当目标变量不满足高斯分布时,其结果可能偏离真正最优.零误差密度函数(ZED)利用非参数估计中的Parzen窗法得到误差在零点的概率密度函数.将零误差密度函数作为BP网络的目标函数时,通过对光滑参数的选择使新的目标函数能够适用于期望输出满足任意分布.仿真实验分别以零误差密度函数和均方误差函数为目标函数的BP网络学习在函数逼近方面进行比较,结果表明零误差密度函数要比均方误差函数的适用范围更广.
|
| 关 键 词： | BP网络 均方误差函数 零误差密度函数 非高斯分布 |
Learning of BP Neural Networks Based on Zero-Error Density Criterion Function |
| |
| ZOU Xiu-ming,YANG Sai,SUN Huai-jiang.Learning of BP Neural Networks Based on Zero-Error Density Criterion Function[J].Journal of Huaiyin Teachers College(Natrual Science Edition),2010,9(4). |
| |
| Authors: | ZOU Xiu-ming YANG Sai SUN Huai-jiang |
| |
| Institution: | ZOU Xiu-ming1,2,YANG Sai2,SUN Huai-jiang2(1.School of Physics and Electronic Electrical Engineering,Huaiyin Normal University,Huaian Jiangsu 223300,China)(2.School of Computer Science and Technology,Nanjing University of Science and Technology,Nanjing Jiangsu 210094,China) |
| |
| Abstract: | BP neural networks usually use mean squares error(MSE) function as the objective function,the results may deviate the optimal values in the condition that expected vectors don't follow Gaussian distribution.zero-error density (ZED) function uses Parzen window method of non-parameter estimation to get error density at origin,which can be used in the condition that expected output vector follow any density distribution by choosing an appropriate smooth parameter. Compared the BP networks with the new cost fun... |
| |
| Keywords: | BP networks mean squared error function zero-error density maximization algorithm non-gaussian distribution |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
