Try answering the following questions:

1. What is the difference between an RNN and a feedforward neural network?
2. How is the hidden state computed in an RNN?
3. What is the use of a recurrent network?
4. How does the vanishing gradient problem occur?
5. What is the exploding gradient problem?
6. How gradient clipping mitigates the exploding gradient problem?
7. What are the different types of RNN architectures?