| 摘要: |
| 对FP算法进行了改进.首先对“共同特征位”进行了优化处理,其次对x0样本的选取进行优化,最后,对阈值设计进行优化处理.经过优化后的样本吸引域明显地较为均匀,模拟实验数据也证明了这一点. |
| 关键词: 神经网络,多层前馈网络,FP算法,共同特征位,吸引域. |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Improvement of the FP Learning and Synthesize Algorithm |
|
WU Ying-quan,WANG Jian-bo,LIU Ming-zhu
|
| Abstract: |
| FP algorithm is improved in this paper. First, the “common characteristic bit” is optimized, then the selection of x0 sample is optimized. Finally, the design of threshold is optimized. After these optimizations, the attractive domain of sample appears more even. The solution is proved by the test data. |
| Key words: Neural network, multilayered feedforward network, FP algorithm, common characteristic bit, attractive domain. |
