How does forget gate prevent LSTM network from vanishing gradient problem?

I am working on LSTM networks (RNN) and read all about LSTM and its different variations. My problem is about the forget gate and its dynamics. I know WHY we use those gates (e.g....

14 March 2017 1,162 2 View