Which are the correct statements about LSTMs$,$ GRUs and RNNs$?$

$$a$)$ LSTMs have fewer parameters to train and require less computational power compared to the GRU.

$$b$)$ LSTMs can remember information for a longer period of time, addressing the vanishing gradient problem.

$$c$)$ The GRU unit has more parameters than the LSTM unit, making them more complex.

$$d$)$ GRU $($Gated Recurrent Unit$)$ networks are known for combining the forget and input gates into a single update gate, simplifying the model architecture.

$$e$)$ In RNNs$,$ vanishing gradient problem occurs when gradients grow exponentially as they are propagated back through time.

$$f$)$ It is a problem for RNNs where gradients become too small, preventing the network from learning long$-$term dependencies.

$$g$)$ Hidden state in RNNs acts as the network's memory, carrying information from one step of the network to the next.