What do you know about the convolutional term in convolutional neural network (CNN)? Choose all correct answers

• It refers to a mathematical operation that produces a high dimensional tensor

• It refers to the mathematical function that slides a filter over the input

• It breaks down the input feature into smaller parts

• The term convolution refers to the mathematical combination of three functions to produce the fourth function

What is the desired output of the CNN layer?

• Kernel

• Feature map

• Decomposed input

• Classifier

Why do we need to put fully-connected layers in the end of CNN in the case of MNIST dataset? Choose the best answer

• CNN only yield feature map

• To improve accuracy even better

• MNIST dataset contains a lot of noise

• The activation function in CNN is not suitable for a classification task

For which purpose Convolutional Neural Network is used?

• It is a multi purpose alghorithm that can be used for Unsupervised Learning.

• Mainly to process and analyse digital images, with some success cases involving processing voice and natural language.

• It is a multi purpose alghorithm that can be used for Reinforced Learning.

• Mainly to process and analyse financial models, predicting future trends.

What is the biggest advantage utilizing CNN?

• Little dependence on pre-processing, decreasing the needs of human effort developing its functionalities.

• It is easy to understand and fast to implement.

• It has the highest accuracy among all algorithms that predicts images.

• It works well both for Supervised and Unsupervised Learning.

Which answer explains better the Pooling?

• Creates a pool of data in order to improve the accuracy of the alghorithm predicting images.

• It assists in the detection of distorted features, in order to find dominant attributes.

• It assists in the detection of features, even if they are distorted, in addition to decreasing the attribute sizes, resulting in decreased computational need.

• Decrease the features size, in order to decrease the computional power that are needed.

What do Shared Weights mean in CNN? Choose the best answer

• It calculates the feature´s weights and compares with other algorithms in order to find the best parameters.

• Sharing weights among the features, make it easier and faster to CNN predict the correct image.

• Forcing the neurons of one layer to share weights, the forward pass becomes the equivalent of convolving a filter over the tensor to produce a new tensor.

• It means that CNN uses the weights of each feature in order to find the best model to make a prediction, sharing the results and returning the average.

What type of cost function is used in a convolutional neural network?

• Categorical cross-entropy

• RMSE

• Maximum Likelihood

• It depends on the task.

Suppose we do convolution upon an input image and a kernel, then what is the output of this operation?

• -3

• -1

• 1

• 3

Which researcher below mostly renowned by his/her work in CNN? Choose the best answer

• Andrew Ng

• Yann LeCun

• Geoffrey Hinton

• Yoshua Bengio