| 摘要: |
| 针对BP神经网络存在易陷入局部极值的缺陷，提出一种基于改进的人工鱼群算法优化的BP神经网络.先用改进的人工鱼群算法优化BP神经网络的初始权值和阀值，然后再执行BP算法训练BP神经网络的权值和阀值.函数拟合仿真实验表明该优化方法提高了BP神经网络的泛化性能. |
| 关键词: BP神经网络 人工鱼群算法 优化 仿真 |
| DOI： |
| 分类号:TP18 |
| 基金项目:湖南省自科基金资助项目（13JJ3083）；湖南省教育厅青年项目(13B020) |
|
| A research on the improvement of artificial fish algorithm and its optimization on BP neural network |
|
Gong Bo,Zeng Feiyan
|
|
School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan 411201,China
|
| Abstract: |
| Aiming at the defect that BP neural network easily falls into a local extremum, an optimized BP neural network was proposed that based on modified artificial fish algorithm. First, the initial weights and threshold were optimized through improving artificial fish algorithm. Then it implemented BP algorithm to train the weights and values of BP neural network. Finally, it is proved that the proposed optimization method improves the generalization performance of BP neural network according to function fitting simulation experiments. |
| Key words: BP neural network AFSA optimization simulation |
