| 基于贝叶斯算法的神经网络优化方法 |
| |
| 引用本文： | 易鸿.基于贝叶斯算法的神经网络优化方法[J].达县师范高等专科学校学报,2010,20(2):37-40. |
| |
| 作者姓名： | 易鸿 |
| |
| 作者单位： | [1]西华大学电气信息学院,四川成都610039 [2]四川文理学院物理与工程技术系,四川达州635000 |
| |
| 摘 要： | 提出了一种近似建模的前馈网络训练算法一贝叶斯算法，该方法能对模型中的未知量构造其后验分布，提高网络的泛化性能，获取对应于后验分布最大值的权值向量．结果表明，贝叶斯算法所建立的神经网络近似模型具有更高、更稳定的精度．
|
| 关 键 词： | 贝叶斯算法 神经网络 遗传算法 |
Optimization Based on Bayesian Neural Network Algorithm |
| |
| YI Hong.Optimization Based on Bayesian Neural Network Algorithm[J].Journal of Daxian Teachers College,2010,20(2):37-40. |
| |
| Authors: | YI Hong |
| |
| Institution: | YI Hong (1. Electrical and Information Engineering School of Xihua University, Chengdu Sichuan 610039, 2. Physics and Engineering Technology Department of SASU, Dazhou Siehuan 635000, China) |
| |
| Abstract: | This paper presents a model for the approximation of the feed -forward network training algorithms： Bayesian algorithm. This methed can model the amount of construction unknown posterior distribution, improve the network generalization performance, access to the posterior distribution corresponding to the maximum weight vector. The results show the approximate model to be established on Bayesian neural network algorithms has a higher and more stable accuracy. |
| |
| Keywords: | Bayesian algorithm neural network genetic algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
