反向传播

• 一些笔记（未完待续）

文章中的英文描述，公式以及图片，均来自吴恩达深度学习课程的课后作业

$\frac{\partial \mathcal{J} }{ \partial z_{2}^{(i)} } = \frac{1}{m} (a^{[2](i)} - y^{(i)})$∂z2(i)​∂J​=m1​(a[2](i)−y(i))

$\frac{\partial \mathcal{J} }{ \partial W_2 } = \frac{\partial \mathcal{J} }{ \partial z_{2}^{(i)} } a^{[1] (i) T}$∂W2​∂J​=∂z2(i)​∂J​a[1](i)T

$\frac{\partial \mathcal{J} }{ \partial b_2 } = \sum_i{\frac{\partial \mathcal{J} }{ \partial z_{2}^{(i)}}}$∂b2​∂J​=∑i​∂z2(i)​∂J​

$\frac{\partial \mathcal{J} }{ \partial z_{1}^{(i)} } = W_2^T \frac{\partial \mathcal{J} }{ \partial z_{2}^{(i)} } * ( 1 - a^{[1] (i) 2})$∂z1(i)​∂J​=W2T​∂z2(i)​∂J​∗(1−a[1](i)2)

$\frac{\partial \mathcal{J} }{ \partial W_1 } = \frac{\partial \mathcal{J} }{ \partial z_{1}^{(i)} } X^T$∂W1​∂J​=∂z1(i)​∂J​XT

$\frac{\partial \mathcal{J} _i }{ \partial b_1 } = \sum_i{\frac{\partial \mathcal{J} }{ \partial z_{1}^{(i)}}}$∂b1​∂Ji​​=∑i​∂z1(i)​∂J​

下图是反向传播时的梯度计算，输出层**函数为Sigmoid函数，隐藏层的**函数是tanh()，右侧是对应的向量化实现。

• Note that $*$∗ denotes elementwise multiplication.

• The notation you will use is common in deep learning coding:

  • dW1 = $\frac{\partial \mathcal{J} }{ \partial W_1 }$∂W1​∂J​
  • db1 = $\frac{\partial \mathcal{J} }{ \partial b_1 }$∂b1​∂J​
  • dW2 = $\frac{\partial \mathcal{J} }{ \partial W_2 }$∂W2​∂J​
  • db2 = $\frac{\partial \mathcal{J} }{ \partial b_2 }$∂b2​∂J​

• Tips:

  • To compute dZ1 you’ll need to compute $g^{[1]'}(Z^{[1]})$g[1]′(Z[1]). Since $g^{[1]}(.)$g[1](.) is the tanh activation function, if $a = g^{[1]}(z)$a=g[1](z) then $g^{[1]'}(z) = 1-a^2$g[1]′(z)=1−a2. So you can compute
    $g^{[1]'}(Z^{[1]})$g[1]′(Z[1]) using (1 - np.power(A1, 2)).