“`” 答:推导forget gate, input gate, cell state, hidden information等的变化;因为LSTM有进有出且当前的cellinformation是通过input gate控制之后叠加的,RNN是叠乘,因此LSTM可以防止梯度消失或者梯度爆炸

 

 

<pre><code> "“`

Was this helpful?

0 / 0

发表回复 0

Your email address will not be published.