Check your understanding of this module before moving on.
?
0/3
Checkpoint quiz 3 questions
A quick check on this module's key ideas before you move on
Question 01
What is the main innovation that lets LSTMs retain information over much longer sequences than vanilla RNNs?
A is correct. LSTMs add a cell state and gating mechanisms (forget, input, output gates) that let the network learn when to preserve, update, or discard information. This gives gradients a much more stable path through the cell state across many time steps.
Question 02
What does the forget gate in an LSTM control?
A is correct. The forget gate outputs a value between 0 and 1 for each element of the cell state, deciding how much of that prior information to keep (close to 1) or erase (close to 0) before incorporating the new input.
Question 03
How does a GRU (Gated Recurrent Unit) compare structurally to an LSTM?
A is correct. GRUs combine the LSTM's separate cell and hidden states into a single state and use just two gates (reset and update) instead of three, giving a simpler architecture that's faster to compute while often matching LSTM performance.