# 神经网络反向传播BP算法

## 1. BP算法说明

• 输出层误差

E = 1 2 ( d − O ) 2 = 1 2 ∑ k = 1 ℓ ( d k − O k ) 2 E=\frac{1}{2}(d-O)^{2}=\frac{1}{2} \sum_{k=1}^{\ell}\left(d_{k}-O_{k}\right)^{2}

• 隐藏层误差

E = 1 2 ∑ k = 1 c ( d k − f ( n e t k ) ) 2 = 1 2 ∑ k = 1 ℓ ( d k − f ( ∑ j = 1 m w j k y j ) ) 2 E=\frac{1}{2} \sum_{k=1}^{c}\left(d_{k}-f\left(n e t_{k}\right)\right)^{2}=\frac{1}{2} \sum_{k=1}^{\ell}\left(d_{k}-f\left(\sum_{j=1}^{m} w_{j k} y_{j}\right)\right)^{2}

• 输入层误差

E = 1 2 ∑ k = 1 c ( d k − f [ ∑ j = 0 m w j k f ( n e t j ) ] ) 2 = 1 2 ∑ k = 1 ℓ ( d k − f [ ∑ j = 0 m w j k f ( ∑ i = 1 n v i j x i ) ] ) 2 E=\frac{1}{2} \sum_{k=1}^{c}\left(d_{k}-f\left[\sum_{j=0}^{m} w_{j k} f\left(n e t_{j}\right)\right]\right)^{2}=\frac{1}{2} \sum_{k=1}^{\ell}\left(d_{k}-f\left[\sum_{j=0}^{m} w_{j k} f\left(\sum_{i=1}^{n} v_{i j} x_{i}\right)\right]\right)^{2}

Δ ω j k = − η ∂ E ∂ ω j k j = 0 , 1 , 2 , … , m ; κ = 1 , 2 , … , ℓ \Delta \omega_{j \mathrm{k}}=-\eta \frac{\partial E}{\partial \omega_{j \mathrm{k}}} j=0,1,2, \ldots, m ; \quad \kappa=1,2, \ldots, \ell

Δ v i j = − η ∂ E ∂ v i j i = 0 , 1 , 2 , … , n ; j = 1 , 2 , … , m \Delta v_{i j}=-\eta \frac{\partial E}{\partial v_{i j}} i=0,1,2, \ldots, n ; \quad j=1,2, \ldots, m

• 点赞
• 评论 1
• 分享
x

海报分享

扫一扫，分享海报

• 收藏
• 打赏

打赏

食得咸鱼抵得渴

你的鼓励将是我创作的最大动力

C币 余额
2C币 4C币 6C币 10C币 20C币 50C币
• 举报
• 一键三连

点赞Mark关注该博主, 随时了解TA的最新博文

07-30 1万+

05-27 8475
05-09 5万+
10-24 1万+
09-29 490
09-16 2万+
11-04 267
04-01 15万+
04-03 383
01-06 1666
10-08 1万+
06-28 1332