论文解惑《word2vec Parameter Learning Explained》1.1--CBOW模型中One-word context情况公式推导问题

  word2vec中有CBOW和Skip-Gram模型，对于两个模型中的参数如何学习的公式推导，在《word2vec Parameter Learning Explained》中有详细解释，我在阅读1.1节One-word context时对于公式(8)的推导感到不解，花了些时间，原文如下：
  “Let us now derive the update equation of the weights between hidden and output layers. Take the derivative of E with regard to $j$-th unit’s net input $u_j$, we obtain$$\frac{\partial E}{\partial u_j}=y_j-t_j:=e_j$$  where $t_j=1(j=j^{\ast }),\text{i.e},t_j$ will only be 1 when the $j$-th unit is the output word, otherwise $t_j=0.$”
  我一开始不明白是怎么推到这一步的，后来发现过程很显然：
$$\begin{array}{rl}E& =\text{log}\sum _{j^{\prime }=1}^V\text{exp}(u_{j^{\prime }})-u_{j\ast }\\ e_j=\frac{\partial E}{\partial u_j}& =\frac{\text{exp}(u_j)}{{\sum }_{j^{\prime }=1}^V\text{exp}(u_{j^{\prime }})}-u_{j\ast }\\ & =y_j-u_{j\ast }\\ & =y_j-t_i\end{array}$$