QUESTION $1$

Music generation using RNNs can be done using a:

Sequence$-$to$-$vector RNN model

Sequence$-$to$-$sequence RNN model

Vector$-$to$-$sequence RNN model

QUESTION $2$

RNNs are prone to exploding $($or vanishing$)$ gradients, which is particularly problematic when dealing with long sequences.

True

False

QUESTION $3$

Text generation using RNNs can be done using a:

Sequence$-$to$-$vector RNN model

Sequence$-$to$-$sequence RNN model

Vector$-$to$-$sequence RNN model

$q,$

QUESTION $4$

In the transformer architecture the most important layer is:

The feed forward layers

The multi$-$attention layer

The normalization layers

$q,$

$q,$ QUESTION $5$

The input dimensions of a RNN layer are: $[$ BatchSize X NumberTimeSteps X NumberinputFeatures$]$

$$ True

$$ False

$q,$

$$

QUESTION $6$

Time series forecasting $($prediction$)$ using RNNs can be done using a:

Sequence$-$to$-$vector RNN model

Sequence$-$to$-$sequence RNN model

Vector$-$to$-$sequence RNN model