KPMG Deep learning CNN and RNN

Which statement best describes the architecture of a basic RNN?

• a) It only has a single input and output layer.

• b) It has multiple hidden layers and a single output layer.

• c) It can have multiple layers of varying types.

• d) It has a feedback connection that allows information to flow from one step to the next.

RNN Stands for

• Recurrent Neural Network

• Recurrent Normal Network

• Register Neural Network

• None of the above

The output which is computed in RNN again send back to Neuron as input

• True

• False

• May be

• None of the above

RNN remembers each and every information through

• Work

• Time

• Hours

• Memory

In RNN Each unit has an internal state which is called the

• visible state of unit

• hidden state of the unit.

• Visible function

• Hidden function

CNN stands for

• Convolutional Neural Network

• Convolutional Near Network

• Convolutional Neural Form

• None of the above

Which of the following activation functions is commonly used in CNNs?

• a) Sigmoid

• b) ReLU (Rectified Linear Unit)

• c) Tanh (Hyperbolic tangent)

• d) All of the above

In a typical CNN architecture, what layer follows the convolutional layer?

• a) Activation layer

• b) Pooling layer

• c) Fully connected layer

• d) Dropout layer

Max pooling in CNN takes the maximum value from the selected part

• True

• False

• May be

• None of the above

The sequence in CNN is

• Input, Convolution, Pooling, Fully connected

• Convolution, Pooling, Fully connected, Input

• Input, Fully connected, Convolution, Pooling,

• None of the above