Neural Networks and NLP Quiz

Which of the following is FALSE about Padding in CNN?

• We should use valid padding if we know that information at edges is not that much useful

• We compromise to lose some edge information of the image in zero padding

• There is no reduction in dimension when we use zero padding

• In valid padding, we drop the part of the image where the filter does not fit

Which of the following is FALSE about zero padding?

• It is used to preserve the spatial size of the input volume

• It is used to preserve edge information of the image

• It is used to preserve resolution of the image

• None of the above

Which of the following is TRUE about Padding in CNN?

• Padding is used in convolution layer as well as in pooling layer

• Padding is used in convolution layer as well as in fully connected layer

• Padding is used in fully connected layer as well as in pooling layer

• Padding is used only in convolution layer

Filter of size 3X3 is rotated over input matrix of size 4X4 (stride=1). What will be the size of output matrix after applying zero padding?

• 4X4

• 3X3

• 2X2

• 1X1

Filter of size 3X3 is rotated over input matrix of size 4X4 (stride=1). What will be the size of output matrix after applying valid padding?

• 4X4

• 3X3

• 2X2

• 1X1

Which of the following is FALSE about Pooling Layer in CNN?

• Pooling layer must be added after each convolutional layer

• Output of convolutional layer acts as an input to the pooling layer

• It does down-sampling of an image which reduces dimensions by retaining vital information

• It does feature extraction and detects components of the image like edges, corners etc.

Which of the following is TRUE about Pooling Layer in CNN?

• We can use Max, Min, Average or Sum pooling in CNN

• It helps in retaining the most useful information and throwing away useless information

• It reduces resolution and dimension and hence reduces computational complexity

• All of the above

Which of the following is FALSE about Pooling Layer in CNN?

• It helps in reducing overfitting

• It reduces computational complexity

• It increases image resolution

• None of the above

Which of the following is NOT a hyper-parameter in CNN?

• Code size for compression

• Number and size of kernels in a convolutional layer

• Padding in a convolutional layer (zero or valid padding)

• Number of convolutional layers

Which of the following model is best suited for sequential data?

• Convolutional Neural Networks (ConvNots)

• Capsule Neural Networks (CapsNots)

• RNN (Recurrent Neural Network)

• Autoencoders